diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..51ef049
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,79 @@
+.ipynb_checkpoints
+lib/build
+lib/pycocotools/_mask.c
+lib/pycocotools/_mask.so
+caffe
+output
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*,cover
+.hypothesis/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# IPython Notebook
+.ipynb_checkpoints
+
+# pyenv
+.python-version
+
+# celery beat schedule file
+celerybeat-schedule
+
+# dotenv
+.env
+
+# virtualenv
+venv/
+ENV/
+
+# Spyder project settings
+.spyderproject
+
+# Rope project settings
+.ropeproject
diff --git a/ImageSets/Main/test.txt b/ImageSets/Main/test.txt
new file mode 100644
index 0000000..30177c1
--- /dev/null
+++ b/ImageSets/Main/test.txt
@@ -0,0 +1,4952 @@
+000001
+000002
+000003
+000004
+000006
+000008
+000010
+000011
+000013
+000014
+000015
+000018
+000022
+000025
+000027
+000028
+000029
+000031
+000037
+000038
+000040
+000043
+000045
+000049
+000053
+000054
+000055
+000056
+000057
+000058
+000059
+000062
+000067
+000068
+000069
+000070
+000071
+000074
+000075
+000076
+000079
+000080
+000082
+000084
+000085
+000086
+000087
+000088
+000090
+000092
+000094
+000096
+000097
+000098
+000100
+000103
+000105
+000106
+000108
+000111
+000114
+000115
+000116
+000119
+000124
+000126
+000127
+000128
+000135
+000136
+000137
+000139
+000144
+000145
+000148
+000149
+000151
+000152
+000155
+000157
+000160
+000166
+000167
+000168
+000172
+000175
+000176
+000178
+000179
+000181
+000182
+000183
+000185
+000186
+000188
+000191
+000195
+000196
+000197
+000199
+000201
+000202
+000204
+000205
+000206
+000212
+000213
+000216
+000217
+000223
+000226
+000227
+000230
+000231
+000234
+000237
+000238
+000239
+000240
+000243
+000247
+000248
+000252
+000253
+000254
+000255
+000258
+000260
+000261
+000264
+000265
+000267
+000271
+000272
+000273
+000274
+000277
+000279
+000280
+000281
+000283
+000284
+000286
+000287
+000290
+000291
+000292
+000293
+000295
+000297
+000299
+000300
+000301
+000309
+000310
+000313
+000314
+000315
+000316
+000319
+000324
+000326
+000327
+000330
+000333
+000335
+000339
+000341
+000342
+000345
+000346
+000348
+000350
+000351
+000353
+000356
+000357
+000358
+000360
+000361
+000362
+000364
+000365
+000366
+000368
+000369
+000371
+000375
+000376
+000377
+000378
+000383
+000384
+000385
+000386
+000388
+000389
+000390
+000392
+000393
+000397
+000398
+000399
+000401
+000402
+000405
+000409
+000410
+000412
+000413
+000414
+000415
+000418
+000421
+000422
+000423
+000425
+000426
+000429
+000432
+000434
+000436
+000437
+000440
+000441
+000442
+000444
+000445
+000447
+000449
+000451
+000452
+000453
+000455
+000456
+000457
+000458
+000465
+000466
+000467
+000471
+000472
+000473
+000475
+000478
+000479
+000481
+000485
+000487
+000488
+000490
+000493
+000495
+000497
+000502
+000504
+000505
+000506
+000507
+000510
+000511
+000512
+000517
+000521
+000527
+000529
+000532
+000533
+000534
+000536
+000538
+000539
+000542
+000546
+000547
+000548
+000551
+000553
+000556
+000557
+000558
+000560
+000561
+000562
+000566
+000567
+000568
+000569
+000570
+000571
+000572
+000573
+000574
+000575
+000576
+000578
+000580
+000584
+000585
+000586
+000587
+000593
+000594
+000595
+000596
+000600
+000602
+000603
+000604
+000606
+000607
+000611
+000614
+000615
+000616
+000617
+000618
+000621
+000623
+000624
+000627
+000629
+000630
+000631
+000634
+000636
+000638
+000639
+000640
+000641
+000642
+000643
+000644
+000646
+000649
+000650
+000651
+000652
+000655
+000658
+000659
+000662
+000664
+000665
+000666
+000668
+000669
+000670
+000673
+000674
+000678
+000679
+000681
+000683
+000687
+000691
+000692
+000693
+000696
+000697
+000698
+000701
+000703
+000704
+000706
+000708
+000715
+000716
+000718
+000719
+000721
+000722
+000723
+000724
+000725
+000727
+000732
+000734
+000735
+000736
+000737
+000741
+000743
+000744
+000745
+000747
+000749
+000751
+000757
+000758
+000759
+000762
+000765
+000766
+000769
+000773
+000775
+000778
+000779
+000781
+000783
+000784
+000785
+000788
+000789
+000790
+000792
+000795
+000798
+000801
+000803
+000807
+000809
+000811
+000813
+000817
+000819
+000821
+000824
+000825
+000833
+000835
+000836
+000837
+000838
+000839
+000840
+000841
+000844
+000846
+000852
+000853
+000856
+000858
+000861
+000864
+000866
+000869
+000870
+000873
+000875
+000877
+000881
+000883
+000884
+000886
+000890
+000891
+000893
+000894
+000897
+000901
+000905
+000907
+000909
+000910
+000913
+000914
+000916
+000922
+000924
+000925
+000927
+000928
+000930
+000932
+000933
+000938
+000939
+000940
+000941
+000942
+000944
+000945
+000952
+000953
+000955
+000956
+000957
+000959
+000960
+000961
+000963
+000968
+000969
+000970
+000974
+000975
+000976
+000978
+000979
+000981
+000983
+000984
+000985
+000986
+000988
+000990
+000992
+000994
+000995
+000998
+001000
+001003
+001005
+001006
+001007
+001013
+001016
+001019
+001020
+001021
+001022
+001023
+001025
+001026
+001029
+001030
+001031
+001032
+001033
+001034
+001035
+001037
+001038
+001039
+001040
+001044
+001046
+001047
+001048
+001049
+001051
+001054
+001055
+001058
+001059
+001063
+001065
+001067
+001070
+001075
+001076
+001080
+001081
+001085
+001086
+001087
+001088
+001089
+001090
+001094
+001095
+001096
+001098
+001099
+001100
+001103
+001105
+001108
+001111
+001114
+001115
+001116
+001117
+001118
+001120
+001122
+001123
+001126
+001128
+001131
+001132
+001133
+001134
+001135
+001138
+001139
+001141
+001146
+001150
+001153
+001155
+001157
+001159
+001162
+001163
+001165
+001167
+001169
+001173
+001177
+001178
+001179
+001180
+001181
+001183
+001188
+001189
+001190
+001193
+001195
+001196
+001197
+001198
+001202
+001208
+001210
+001213
+001216
+001217
+001218
+001219
+001220
+001222
+001223
+001227
+001228
+001232
+001235
+001238
+001242
+001243
+001244
+001245
+001246
+001249
+001251
+001252
+001253
+001255
+001256
+001257
+001261
+001262
+001264
+001267
+001271
+001275
+001276
+001278
+001280
+001282
+001283
+001285
+001291
+001295
+001296
+001297
+001300
+001301
+001302
+001303
+001305
+001306
+001307
+001308
+001313
+001317
+001318
+001319
+001320
+001321
+001322
+001328
+001329
+001331
+001335
+001336
+001338
+001339
+001340
+001342
+001344
+001347
+001349
+001351
+001353
+001354
+001355
+001356
+001357
+001358
+001359
+001363
+001366
+001367
+001368
+001369
+001370
+001372
+001373
+001374
+001376
+001377
+001379
+001380
+001381
+001382
+001389
+001391
+001392
+001394
+001396
+001398
+001399
+001401
+001403
+001407
+001410
+001411
+001412
+001415
+001416
+001417
+001419
+001422
+001423
+001424
+001425
+001428
+001429
+001431
+001433
+001435
+001437
+001438
+001440
+001446
+001447
+001448
+001449
+001452
+001454
+001456
+001458
+001459
+001461
+001462
+001469
+001471
+001473
+001474
+001476
+001477
+001478
+001482
+001487
+001489
+001491
+001495
+001496
+001500
+001502
+001503
+001505
+001506
+001507
+001508
+001511
+001513
+001516
+001518
+001519
+001520
+001525
+001527
+001530
+001533
+001534
+001535
+001538
+001540
+001542
+001546
+001547
+001549
+001550
+001551
+001552
+001558
+001560
+001562
+001564
+001566
+001567
+001568
+001569
+001570
+001572
+001573
+001574
+001575
+001578
+001581
+001583
+001584
+001585
+001587
+001589
+001591
+001592
+001596
+001599
+001600
+001601
+001602
+001605
+001606
+001609
+001613
+001615
+001616
+001619
+001620
+001621
+001623
+001624
+001625
+001626
+001629
+001631
+001634
+001635
+001637
+001639
+001641
+001644
+001645
+001646
+001648
+001652
+001655
+001656
+001657
+001658
+001659
+001660
+001663
+001664
+001665
+001666
+001667
+001668
+001670
+001671
+001672
+001674
+001679
+001681
+001687
+001692
+001694
+001695
+001696
+001697
+001698
+001700
+001701
+001702
+001703
+001704
+001705
+001706
+001709
+001710
+001712
+001715
+001716
+001719
+001720
+001722
+001728
+001731
+001735
+001736
+001737
+001740
+001742
+001743
+001744
+001745
+001748
+001751
+001753
+001757
+001760
+001762
+001763
+001764
+001767
+001769
+001770
+001773
+001774
+001776
+001779
+001781
+001783
+001786
+001788
+001790
+001791
+001792
+001794
+001796
+001798
+001802
+001803
+001804
+001805
+001808
+001811
+001812
+001813
+001814
+001815
+001817
+001819
+001820
+001822
+001823
+001824
+001826
+001829
+001831
+001835
+001838
+001839
+001844
+001846
+001848
+001850
+001851
+001852
+001856
+001857
+001859
+001863
+001865
+001866
+001867
+001868
+001869
+001871
+001873
+001874
+001876
+001879
+001880
+001883
+001884
+001885
+001886
+001889
+001890
+001891
+001893
+001895
+001897
+001900
+001905
+001908
+001909
+001910
+001912
+001913
+001914
+001916
+001917
+001919
+001921
+001923
+001924
+001925
+001926
+001929
+001935
+001939
+001942
+001943
+001946
+001947
+001949
+001951
+001953
+001955
+001956
+001957
+001959
+001961
+001965
+001966
+001967
+001968
+001969
+001973
+001974
+001975
+001979
+001983
+001984
+001986
+001987
+001988
+001990
+001991
+001992
+001993
+001994
+001996
+001997
+001998
+002003
+002005
+002007
+002008
+002009
+002010
+002013
+002014
+002016
+002017
+002018
+002026
+002028
+002029
+002031
+002032
+002033
+002035
+002038
+002040
+002041
+002044
+002046
+002048
+002050
+002052
+002053
+002057
+002059
+002060
+002062
+002065
+002066
+002071
+002072
+002073
+002074
+002075
+002076
+002077
+002078
+002079
+002080
+002081
+002084
+002085
+002087
+002089
+002092
+002093
+002097
+002100
+002103
+002105
+002106
+002107
+002110
+002111
+002113
+002115
+002118
+002119
+002121
+002122
+002123
+002127
+002128
+002130
+002131
+002133
+002137
+002138
+002141
+002143
+002144
+002147
+002148
+002149
+002150
+002154
+002157
+002159
+002160
+002161
+002162
+002164
+002167
+002168
+002173
+002175
+002177
+002185
+002188
+002189
+002195
+002198
+002200
+002203
+002204
+002205
+002206
+002207
+002210
+002211
+002216
+002217
+002222
+002223
+002225
+002227
+002229
+002230
+002231
+002232
+002235
+002236
+002239
+002240
+002242
+002243
+002245
+002246
+002250
+002252
+002254
+002258
+002262
+002264
+002269
+002271
+002274
+002275
+002282
+002283
+002286
+002289
+002292
+002294
+002295
+002296
+002297
+002298
+002299
+002301
+002303
+002304
+002309
+002312
+002313
+002314
+002316
+002317
+002319
+002322
+002325
+002326
+002327
+002331
+002336
+002338
+002339
+002341
+002344
+002346
+002349
+002351
+002353
+002356
+002357
+002358
+002360
+002363
+002365
+002370
+002379
+002380
+002381
+002383
+002386
+002388
+002389
+002390
+002394
+002395
+002396
+002397
+002398
+002399
+002400
+002402
+002406
+002408
+002409
+002412
+002414
+002416
+002418
+002421
+002422
+002424
+002426
+002428
+002429
+002430
+002431
+002432
+002434
+002438
+002440
+002446
+002447
+002449
+002451
+002453
+002455
+002457
+002463
+002464
+002467
+002469
+002473
+002474
+002475
+002482
+002484
+002485
+002486
+002487
+002488
+002489
+002495
+002498
+002499
+002503
+002506
+002507
+002509
+002510
+002511
+002515
+002516
+002517
+002521
+002522
+002526
+002527
+002528
+002530
+002531
+002532
+002535
+002536
+002538
+002541
+002543
+002548
+002550
+002551
+002552
+002553
+002556
+002557
+002560
+002562
+002568
+002570
+002573
+002574
+002575
+002576
+002577
+002580
+002581
+002582
+002583
+002587
+002588
+002591
+002592
+002596
+002597
+002601
+002602
+002604
+002607
+002608
+002610
+002612
+002614
+002616
+002617
+002619
+002620
+002622
+002623
+002624
+002626
+002628
+002629
+002630
+002631
+002638
+002639
+002640
+002642
+002644
+002650
+002651
+002652
+002654
+002655
+002656
+002660
+002661
+002663
+002665
+002671
+002672
+002673
+002674
+002676
+002679
+002681
+002685
+002686
+002687
+002688
+002692
+002694
+002698
+002700
+002701
+002703
+002705
+002707
+002708
+002711
+002712
+002716
+002719
+002720
+002724
+002725
+002726
+002728
+002729
+002731
+002733
+002736
+002739
+002740
+002742
+002743
+002746
+002748
+002750
+002752
+002753
+002754
+002756
+002758
+002761
+002764
+002768
+002769
+002770
+002771
+002773
+002777
+002780
+002781
+002787
+002788
+002789
+002790
+002792
+002793
+002797
+002799
+002802
+002805
+002806
+002808
+002809
+002811
+002813
+002814
+002818
+002819
+002821
+002822
+002823
+002824
+002825
+002828
+002829
+002830
+002831
+002832
+002837
+002839
+002840
+002843
+002846
+002849
+002850
+002851
+002852
+002853
+002856
+002857
+002860
+002861
+002862
+002863
+002865
+002871
+002872
+002874
+002876
+002877
+002878
+002882
+002883
+002885
+002887
+002888
+002890
+002892
+002894
+002895
+002897
+002898
+002900
+002902
+002903
+002904
+002905
+002907
+002908
+002909
+002911
+002918
+002920
+002921
+002922
+002923
+002925
+002926
+002927
+002928
+002929
+002930
+002936
+002945
+002948
+002949
+002950
+002951
+002955
+002959
+002961
+002964
+002968
+002970
+002971
+002972
+002973
+002974
+002979
+002980
+002981
+002982
+002983
+002985
+002991
+002993
+002996
+002997
+002998
+002999
+003001
+003006
+003010
+003012
+003014
+003016
+003018
+003019
+003020
+003022
+003025
+003026
+003029
+003030
+003033
+003035
+003036
+003037
+003040
+003041
+003043
+003046
+003048
+003049
+003050
+003052
+003055
+003059
+003060
+003062
+003067
+003068
+003069
+003070
+003071
+003073
+003075
+003076
+003079
+003080
+003081
+003084
+003087
+003091
+003095
+003096
+003097
+003099
+003101
+003104
+003109
+003111
+003113
+003114
+003115
+003119
+003123
+003125
+003128
+003130
+003131
+003132
+003136
+003139
+003141
+003143
+003144
+003148
+003151
+003152
+003153
+003156
+003158
+003160
+003166
+003167
+003168
+003171
+003172
+003173
+003174
+003179
+003180
+003182
+003187
+003190
+003191
+003192
+003193
+003196
+003197
+003198
+003201
+003203
+003206
+003208
+003209
+003212
+003215
+003217
+003220
+003221
+003222
+003224
+003225
+003226
+003227
+003230
+003232
+003234
+003235
+003237
+003238
+003241
+003245
+003246
+003248
+003249
+003251
+003252
+003257
+003263
+003264
+003265
+003266
+003267
+003268
+003275
+003276
+003277
+003278
+003281
+003283
+003286
+003287
+003288
+003289
+003291
+003295
+003297
+003298
+003302
+003304
+003305
+003306
+003309
+003310
+003312
+003314
+003315
+003317
+003318
+003319
+003321
+003322
+003323
+003324
+003326
+003328
+003329
+003332
+003333
+003334
+003340
+003341
+003342
+003345
+003346
+003347
+003348
+003352
+003353
+003357
+003358
+003361
+003364
+003366
+003368
+003371
+003372
+003375
+003378
+003381
+003383
+003384
+003385
+003387
+003388
+003389
+003393
+003394
+003399
+003400
+003402
+003405
+003409
+003411
+003414
+003418
+003423
+003426
+003427
+003428
+003431
+003432
+003434
+003437
+003438
+003440
+003442
+003445
+003446
+003447
+003448
+003454
+003456
+003457
+003459
+003460
+003463
+003467
+003471
+003472
+003473
+003474
+003475
+003476
+003478
+003479
+003480
+003481
+003482
+003483
+003485
+003486
+003488
+003490
+003494
+003495
+003498
+003501
+003502
+003503
+003504
+003505
+003507
+003512
+003513
+003514
+003515
+003517
+003520
+003523
+003526
+003527
+003531
+003532
+003533
+003534
+003535
+003538
+003540
+003541
+003542
+003543
+003544
+003545
+003547
+003552
+003553
+003557
+003558
+003559
+003560
+003561
+003562
+003563
+003568
+003569
+003570
+003571
+003572
+003573
+003574
+003578
+003579
+003581
+003582
+003583
+003584
+003590
+003591
+003592
+003595
+003598
+003600
+003601
+003602
+003607
+003610
+003612
+003613
+003615
+003616
+003617
+003619
+003624
+003626
+003630
+003631
+003633
+003637
+003641
+003643
+003647
+003649
+003650
+003652
+003653
+003659
+003661
+003665
+003666
+003668
+003670
+003672
+003676
+003677
+003680
+003682
+003683
+003686
+003687
+003689
+003692
+003693
+003697
+003701
+003702
+003707
+003710
+003712
+003715
+003716
+003718
+003719
+003720
+003723
+003724
+003725
+003726
+003728
+003730
+003731
+003733
+003734
+003736
+003737
+003738
+003739
+003741
+003742
+003744
+003745
+003746
+003747
+003755
+003756
+003757
+003761
+003762
+003764
+003765
+003766
+003768
+003769
+003770
+003771
+003775
+003776
+003777
+003778
+003782
+003785
+003787
+003789
+003794
+003795
+003799
+003800
+003801
+003802
+003804
+003805
+003810
+003812
+003813
+003815
+003816
+003819
+003822
+003823
+003825
+003829
+003831
+003832
+003833
+003836
+003839
+003840
+003841
+003842
+003843
+003850
+003851
+003852
+003853
+003854
+003858
+003862
+003864
+003867
+003870
+003873
+003875
+003878
+003880
+003881
+003882
+003883
+003884
+003888
+003892
+003893
+003894
+003896
+003897
+003900
+003901
+003902
+003903
+003904
+003906
+003908
+003909
+003910
+003914
+003916
+003917
+003920
+003922
+003925
+003927
+003928
+003929
+003930
+003931
+003933
+003934
+003938
+003940
+003942
+003943
+003944
+003950
+003951
+003952
+003955
+003958
+003959
+003962
+003964
+003967
+003968
+003972
+003975
+003976
+003977
+003978
+003980
+003981
+003982
+003985
+003989
+003995
+003999
+004000
+004001
+004002
+004004
+004006
+004007
+004018
+004021
+004022
+004024
+004026
+004027
+004029
+004030
+004032
+004036
+004038
+004040
+004041
+004042
+004043
+004044
+004045
+004048
+004049
+004050
+004053
+004054
+004055
+004056
+004059
+004061
+004062
+004063
+004064
+004065
+004068
+004070
+004071
+004072
+004074
+004078
+004079
+004080
+004081
+004083
+004084
+004086
+004088
+004090
+004094
+004096
+004097
+004098
+004099
+004101
+004103
+004104
+004107
+004109
+004112
+004114
+004115
+004116
+004118
+004119
+004123
+004124
+004125
+004126
+004127
+004128
+004130
+004132
+004134
+004139
+004144
+004147
+004151
+004153
+004154
+004155
+004156
+004157
+004159
+004160
+004161
+004162
+004165
+004166
+004167
+004172
+004173
+004175
+004176
+004177
+004179
+004180
+004181
+004182
+004183
+004184
+004187
+004188
+004197
+004198
+004199
+004202
+004206
+004207
+004208
+004210
+004211
+004213
+004214
+004216
+004217
+004218
+004219
+004222
+004225
+004226
+004227
+004233
+004234
+004235
+004236
+004238
+004240
+004243
+004245
+004248
+004249
+004250
+004251
+004252
+004254
+004260
+004261
+004262
+004266
+004267
+004268
+004276
+004277
+004278
+004282
+004285
+004288
+004289
+004290
+004294
+004297
+004299
+004301
+004302
+004305
+004306
+004308
+004309
+004311
+004313
+004314
+004316
+004317
+004319
+004320
+004324
+004328
+004330
+004332
+004334
+004335
+004336
+004337
+004340
+004342
+004343
+004344
+004348
+004350
+004353
+004355
+004357
+004358
+004362
+004363
+004366
+004373
+004374
+004375
+004377
+004378
+004381
+004382
+004383
+004385
+004388
+004393
+004394
+004395
+004398
+004399
+004400
+004401
+004402
+004403
+004406
+004407
+004408
+004410
+004412
+004413
+004414
+004415
+004416
+004417
+004418
+004419
+004420
+004422
+004425
+004426
+004427
+004428
+004431
+004435
+004440
+004442
+004443
+004444
+004445
+004447
+004448
+004449
+004451
+004453
+004454
+004456
+004458
+004460
+004461
+004462
+004465
+004467
+004469
+004472
+004473
+004475
+004476
+004477
+004478
+004480
+004482
+004483
+004485
+004486
+004489
+004491
+004492
+004497
+004501
+004503
+004504
+004505
+004506
+004511
+004513
+004515
+004516
+004521
+004522
+004523
+004525
+004529
+004531
+004533
+004534
+004536
+004538
+004541
+004543
+004545
+004546
+004547
+004550
+004554
+004556
+004557
+004559
+004560
+004561
+004564
+004567
+004568
+004569
+004572
+004573
+004575
+004577
+004578
+004580
+004582
+004583
+004586
+004589
+004590
+004593
+004594
+004596
+004598
+004599
+004602
+004603
+004608
+004610
+004613
+004614
+004615
+004616
+004617
+004619
+004620
+004621
+004624
+004629
+004633
+004635
+004637
+004638
+004639
+004640
+004641
+004642
+004645
+004646
+004650
+004657
+004658
+004659
+004661
+004663
+004664
+004665
+004666
+004667
+004668
+004669
+004670
+004677
+004678
+004680
+004681
+004684
+004688
+004690
+004695
+004696
+004697
+004698
+004700
+004703
+004704
+004709
+004711
+004712
+004713
+004716
+004717
+004720
+004721
+004724
+004725
+004726
+004728
+004729
+004730
+004731
+004733
+004734
+004736
+004738
+004739
+004740
+004741
+004744
+004745
+004749
+004751
+004752
+004755
+004756
+004757
+004758
+004759
+004762
+004763
+004764
+004765
+004766
+004767
+004769
+004771
+004772
+004774
+004775
+004778
+004780
+004781
+004784
+004787
+004791
+004795
+004798
+004800
+004802
+004803
+004804
+004806
+004807
+004809
+004810
+004811
+004813
+004817
+004819
+004820
+004821
+004822
+004824
+004827
+004829
+004833
+004835
+004838
+004843
+004844
+004845
+004847
+004851
+004853
+004854
+004855
+004858
+004860
+004861
+004862
+004864
+004865
+004870
+004871
+004874
+004875
+004877
+004880
+004881
+004883
+004884
+004887
+004888
+004889
+004891
+004892
+004893
+004894
+004899
+004900
+004901
+004904
+004906
+004908
+004909
+004914
+004915
+004917
+004918
+004919
+004920
+004921
+004922
+004923
+004924
+004925
+004927
+004930
+004932
+004933
+004934
+004937
+004940
+004941
+004942
+004944
+004945
+004947
+004949
+004952
+004957
+004959
+004964
+004965
+004969
+004970
+004971
+004975
+004978
+004979
+004980
+004981
+004988
+004989
+004993
+004996
+005000
+005002
+005005
+005008
+005009
+005010
+005011
+005012
+005013
+005015
+005017
+005019
+005021
+005022
+005025
+005030
+005031
+005034
+005035
+005038
+005040
+005041
+005043
+005044
+005046
+005048
+005049
+005050
+005051
+005053
+005059
+005060
+005066
+005069
+005070
+005074
+005075
+005076
+005080
+005082
+005083
+005087
+005088
+005089
+005091
+005092
+005095
+005096
+005098
+005099
+005100
+005103
+005105
+005106
+005109
+005112
+005113
+005115
+005117
+005118
+005119
+005120
+005123
+005125
+005126
+005127
+005132
+005133
+005137
+005139
+005140
+005141
+005142
+005147
+005148
+005149
+005151
+005152
+005154
+005155
+005157
+005158
+005162
+005163
+005164
+005165
+005166
+005167
+005170
+005172
+005174
+005178
+005180
+005182
+005184
+005187
+005188
+005192
+005193
+005194
+005196
+005197
+005198
+005200
+005201
+005204
+005205
+005206
+005207
+005211
+005213
+005216
+005218
+005221
+005225
+005226
+005227
+005228
+005232
+005233
+005234
+005235
+005237
+005238
+005240
+005241
+005243
+005247
+005249
+005250
+005251
+005252
+005255
+005256
+005261
+005265
+005266
+005271
+005272
+005275
+005276
+005277
+005279
+005280
+005282
+005284
+005286
+005287
+005289
+005291
+005294
+005295
+005296
+005299
+005300
+005301
+005302
+005308
+005309
+005313
+005316
+005317
+005321
+005322
+005323
+005324
+005329
+005330
+005332
+005333
+005334
+005335
+005339
+005341
+005342
+005347
+005353
+005354
+005356
+005357
+005359
+005361
+005362
+005364
+005366
+005372
+005375
+005376
+005377
+005381
+005382
+005386
+005390
+005392
+005394
+005399
+005400
+005401
+005402
+005403
+005409
+005411
+005412
+005415
+005422
+005425
+005426
+005427
+005428
+005432
+005435
+005437
+005442
+005443
+005444
+005446
+005447
+005449
+005452
+005456
+005458
+005459
+005460
+005462
+005463
+005464
+005466
+005468
+005472
+005473
+005474
+005476
+005477
+005479
+005480
+005482
+005484
+005488
+005490
+005491
+005492
+005493
+005494
+005495
+005498
+005500
+005501
+005502
+005503
+005504
+005505
+005506
+005512
+005513
+005516
+005520
+005523
+005525
+005528
+005529
+005532
+005533
+005534
+005537
+005538
+005540
+005543
+005545
+005546
+005548
+005551
+005553
+005555
+005556
+005557
+005558
+005560
+005561
+005562
+005564
+005565
+005567
+005569
+005570
+005571
+005572
+005575
+005578
+005580
+005581
+005587
+005589
+005594
+005595
+005596
+005597
+005598
+005602
+005604
+005607
+005610
+005612
+005616
+005617
+005619
+005621
+005622
+005623
+005626
+005627
+005628
+005632
+005633
+005634
+005635
+005638
+005642
+005643
+005646
+005649
+005650
+005651
+005656
+005659
+005661
+005663
+005665
+005666
+005667
+005670
+005671
+005673
+005675
+005677
+005678
+005681
+005683
+005684
+005688
+005689
+005690
+005691
+005692
+005694
+005698
+005703
+005706
+005707
+005708
+005709
+005711
+005712
+005717
+005720
+005721
+005722
+005724
+005725
+005726
+005727
+005733
+005734
+005737
+005739
+005744
+005745
+005746
+005748
+005750
+005751
+005753
+005754
+005758
+005759
+005763
+005766
+005767
+005770
+005771
+005772
+005774
+005775
+005776
+005777
+005778
+005785
+005787
+005792
+005793
+005795
+005797
+005798
+005800
+005801
+005802
+005804
+005807
+005808
+005809
+005810
+005816
+005820
+005822
+005823
+005827
+005832
+005833
+005834
+005835
+005837
+005842
+005844
+005846
+005847
+005848
+005849
+005855
+005857
+005858
+005862
+005865
+005866
+005869
+005870
+005871
+005872
+005876
+005880
+005882
+005883
+005886
+005887
+005890
+005891
+005892
+005896
+005898
+005900
+005902
+005904
+005907
+005913
+005915
+005916
+005921
+005922
+005924
+005925
+005926
+005927
+005929
+005931
+005932
+005933
+005934
+005935
+005936
+005937
+005939
+005941
+005942
+005943
+005944
+005945
+005946
+005949
+005950
+005953
+005955
+005957
+005958
+005959
+005962
+005965
+005966
+005967
+005969
+005972
+005973
+005974
+005976
+005977
+005978
+005982
+005986
+005987
+005993
+005994
+005997
+005999
+006002
+006003
+006006
+006007
+006008
+006010
+006013
+006014
+006015
+006016
+006017
+006019
+006021
+006022
+006024
+006031
+006032
+006034
+006036
+006037
+006039
+006040
+006044
+006047
+006048
+006049
+006050
+006051
+006052
+006053
+006054
+006056
+006057
+006059
+006060
+006063
+006064
+006068
+006072
+006075
+006076
+006077
+006080
+006081
+006082
+006083
+006085
+006086
+006087
+006090
+006092
+006093
+006094
+006099
+006101
+006102
+006106
+006109
+006110
+006112
+006113
+006114
+006115
+006116
+006118
+006119
+006121
+006122
+006126
+006127
+006132
+006137
+006138
+006142
+006143
+006144
+006145
+006147
+006149
+006152
+006154
+006155
+006157
+006160
+006164
+006165
+006167
+006168
+006169
+006173
+006178
+006182
+006186
+006191
+006192
+006193
+006194
+006195
+006197
+006199
+006200
+006204
+006205
+006207
+006211
+006213
+006217
+006226
+006227
+006228
+006231
+006232
+006237
+006239
+006242
+006244
+006245
+006246
+006248
+006253
+006255
+006256
+006257
+006263
+006265
+006266
+006268
+006271
+006273
+006274
+006278
+006280
+006283
+006287
+006288
+006292
+006293
+006294
+006297
+006298
+006302
+006303
+006307
+006308
+006310
+006311
+006312
+006313
+006315
+006316
+006317
+006322
+006324
+006326
+006327
+006328
+006331
+006332
+006333
+006334
+006336
+006340
+006342
+006343
+006345
+006347
+006354
+006356
+006358
+006359
+006360
+006361
+006364
+006365
+006368
+006370
+006372
+006373
+006376
+006378
+006379
+006380
+006383
+006384
+006386
+006388
+006389
+006390
+006393
+006394
+006397
+006399
+006401
+006402
+006403
+006405
+006406
+006407
+006408
+006410
+006412
+006413
+006414
+006415
+006416
+006420
+006422
+006423
+006426
+006431
+006432
+006435
+006439
+006441
+006446
+006451
+006452
+006453
+006454
+006457
+006460
+006461
+006464
+006467
+006469
+006471
+006477
+006478
+006479
+006481
+006485
+006487
+006489
+006490
+006491
+006493
+006494
+006496
+006498
+006500
+006502
+006504
+006505
+006508
+006510
+006511
+006513
+006514
+006516
+006517
+006518
+006521
+006522
+006525
+006526
+006527
+006528
+006531
+006533
+006535
+006537
+006539
+006540
+006541
+006544
+006545
+006546
+006552
+006554
+006555
+006557
+006558
+006559
+006561
+006563
+006566
+006567
+006568
+006571
+006573
+006574
+006577
+006579
+006580
+006581
+006582
+006586
+006589
+006590
+006591
+006592
+006594
+006596
+006598
+006600
+006601
+006604
+006607
+006608
+006613
+006614
+006615
+006616
+006620
+006623
+006624
+006629
+006630
+006633
+006634
+006639
+006640
+006641
+006642
+006644
+006646
+006649
+006650
+006651
+006653
+006655
+006656
+006659
+006662
+006663
+006665
+006669
+006672
+006675
+006676
+006680
+006683
+006685
+006686
+006688
+006691
+006692
+006693
+006700
+006701
+006705
+006710
+006711
+006712
+006713
+006715
+006716
+006717
+006720
+006721
+006723
+006724
+006728
+006729
+006732
+006733
+006737
+006741
+006742
+006743
+006744
+006745
+006746
+006749
+006750
+006752
+006754
+006756
+006757
+006758
+006763
+006764
+006767
+006770
+006771
+006774
+006775
+006776
+006778
+006779
+006780
+006785
+006787
+006788
+006790
+006791
+006792
+006793
+006795
+006796
+006798
+006801
+006804
+006807
+006809
+006811
+006812
+006815
+006816
+006817
+006818
+006820
+006823
+006826
+006830
+006831
+006832
+006834
+006837
+006843
+006846
+006851
+006853
+006854
+006856
+006857
+006861
+006863
+006870
+006871
+006872
+006873
+006875
+006877
+006879
+006881
+006882
+006885
+006888
+006889
+006890
+006891
+006894
+006895
+006897
+006898
+006901
+006902
+006904
+006905
+006906
+006907
+006913
+006915
+006920
+006921
+006923
+006925
+006926
+006927
+006928
+006929
+006936
+006937
+006938
+006941
+006942
+006946
+006951
+006954
+006955
+006957
+006960
+006961
+006964
+006967
+006969
+006970
+006973
+006974
+006975
+006977
+006978
+006979
+006980
+006982
+006984
+006985
+006986
+006991
+006992
+006993
+006996
+006997
+006998
+006999
+007000
+007001
+007005
+007010
+007012
+007013
+007014
+007015
+007017
+007019
+007024
+007026
+007027
+007028
+007030
+007032
+007034
+007037
+007041
+007043
+007044
+007047
+007051
+007053
+007055
+007057
+007060
+007061
+007063
+007066
+007067
+007069
+007076
+007081
+007082
+007083
+007085
+007087
+007091
+007094
+007096
+007098
+007099
+007102
+007103
+007106
+007107
+007110
+007111
+007112
+007115
+007116
+007118
+007119
+007120
+007124
+007126
+007127
+007131
+007134
+007135
+007136
+007137
+007142
+007143
+007145
+007151
+007155
+007156
+007157
+007158
+007160
+007161
+007164
+007169
+007170
+007171
+007173
+007175
+007176
+007178
+007179
+007181
+007183
+007186
+007188
+007190
+007192
+007195
+007196
+007198
+007199
+007201
+007202
+007203
+007206
+007207
+007209
+007218
+007220
+007221
+007225
+007226
+007228
+007229
+007231
+007232
+007233
+007235
+007237
+007238
+007239
+007240
+007242
+007246
+007248
+007251
+007252
+007253
+007254
+007255
+007257
+007262
+007264
+007265
+007267
+007268
+007269
+007272
+007273
+007277
+007278
+007281
+007282
+007286
+007287
+007288
+007290
+007291
+007293
+007301
+007303
+007304
+007306
+007307
+007309
+007310
+007312
+007313
+007315
+007316
+007317
+007319
+007320
+007321
+007324
+007326
+007328
+007331
+007332
+007333
+007335
+007337
+007338
+007339
+007340
+007341
+007342
+007345
+007347
+007348
+007349
+007352
+007353
+007354
+007355
+007357
+007358
+007360
+007362
+007364
+007366
+007367
+007368
+007371
+007377
+007378
+007379
+007380
+007382
+007384
+007386
+007387
+007391
+007392
+007393
+007395
+007397
+007399
+007400
+007401
+007402
+007403
+007404
+007405
+007406
+007407
+007409
+007412
+007415
+007418
+007420
+007423
+007426
+007428
+007429
+007430
+007434
+007440
+007441
+007442
+007444
+007447
+007450
+007452
+007453
+007455
+007456
+007459
+007462
+007463
+007464
+007469
+007471
+007472
+007473
+007476
+007478
+007485
+007487
+007488
+007492
+007494
+007495
+007496
+007499
+007500
+007501
+007502
+007504
+007505
+007507
+007508
+007509
+007510
+007512
+007514
+007515
+007516
+007518
+007520
+007522
+007529
+007531
+007532
+007534
+007539
+007541
+007542
+007545
+007548
+007549
+007550
+007552
+007553
+007554
+007556
+007557
+007560
+007561
+007562
+007564
+007567
+007569
+007573
+007574
+007577
+007580
+007581
+007582
+007583
+007584
+007587
+007588
+007589
+007591
+007593
+007595
+007596
+007597
+007598
+007599
+007602
+007604
+007607
+007608
+007609
+007610
+007613
+007616
+007617
+007620
+007623
+007625
+007627
+007628
+007630
+007632
+007634
+007635
+007636
+007638
+007641
+007643
+007644
+007645
+007646
+007648
+007651
+007652
+007658
+007659
+007660
+007661
+007665
+007669
+007674
+007676
+007681
+007684
+007686
+007689
+007690
+007693
+007695
+007698
+007700
+007701
+007703
+007706
+007707
+007708
+007710
+007711
+007714
+007716
+007717
+007719
+007722
+007725
+007726
+007728
+007730
+007733
+007734
+007737
+007738
+007739
+007741
+007744
+007747
+007750
+007752
+007755
+007756
+007757
+007759
+007761
+007764
+007766
+007769
+007770
+007771
+007774
+007778
+007780
+007782
+007783
+007784
+007785
+007787
+007788
+007789
+007792
+007794
+007796
+007797
+007800
+007801
+007802
+007804
+007805
+007806
+007807
+007808
+007811
+007816
+007817
+007818
+007822
+007823
+007825
+007827
+007828
+007829
+007830
+007832
+007835
+007837
+007839
+007842
+007844
+007846
+007848
+007849
+007850
+007851
+007852
+007858
+007860
+007861
+007862
+007866
+007867
+007870
+007871
+007874
+007875
+007879
+007880
+007881
+007882
+007887
+007888
+007891
+007892
+007893
+007894
+007895
+007896
+007903
+007904
+007906
+007907
+007912
+007913
+007917
+007918
+007922
+007927
+007929
+007930
+007934
+007936
+007937
+007938
+007941
+007942
+007944
+007945
+007948
+007949
+007951
+007952
+007955
+007957
+007960
+007961
+007962
+007965
+007966
+007967
+007969
+007972
+007973
+007975
+007977
+007978
+007981
+007982
+007983
+007985
+007986
+007988
+007989
+007990
+007992
+007993
+007994
+007995
+008000
+008003
+008006
+008007
+008010
+008011
+008013
+008014
+008015
+008016
+008018
+008020
+008021
+008022
+008025
+008027
+008028
+008030
+008034
+008035
+008038
+008039
+008041
+008045
+008046
+008047
+008050
+008052
+008054
+008055
+008056
+008058
+008059
+008065
+008066
+008070
+008071
+008073
+008074
+008077
+008078
+008080
+008081
+008088
+008089
+008090
+008092
+008094
+008097
+008099
+008102
+008104
+008109
+008110
+008111
+008113
+008114
+008118
+008119
+008120
+008123
+008124
+008126
+008128
+008129
+008131
+008133
+008134
+008135
+008136
+008143
+008145
+008146
+008147
+008148
+008149
+008152
+008153
+008154
+008155
+008156
+008157
+008158
+008161
+008162
+008165
+008167
+008170
+008172
+008176
+008178
+008179
+008181
+008182
+008183
+008184
+008185
+008187
+008192
+008193
+008194
+008195
+008196
+008198
+008201
+008205
+008206
+008207
+008210
+008212
+008214
+008215
+008217
+008219
+008221
+008227
+008228
+008230
+008231
+008233
+008234
+008237
+008238
+008239
+008240
+008242
+008243
+008245
+008246
+008247
+008249
+008255
+008256
+008257
+008259
+008264
+008265
+008266
+008267
+008270
+008271
+008273
+008274
+008276
+008277
+008278
+008283
+008286
+008287
+008288
+008289
+008290
+008291
+008298
+008303
+008304
+008305
+008308
+008309
+008314
+008321
+008324
+008325
+008328
+008330
+008331
+008333
+008334
+008337
+008339
+008340
+008343
+008344
+008347
+008348
+008350
+008352
+008353
+008354
+008356
+008357
+008358
+008361
+008362
+008363
+008366
+008367
+008369
+008371
+008373
+008375
+008377
+008378
+008379
+008380
+008382
+008383
+008389
+008392
+008393
+008394
+008395
+008396
+008399
+008400
+008401
+008402
+008404
+008405
+008406
+008407
+008408
+008411
+008412
+008414
+008417
+008418
+008419
+008420
+008421
+008428
+008431
+008432
+008435
+008436
+008439
+008440
+008441
+008446
+008447
+008448
+008451
+008455
+008457
+008458
+008459
+008460
+008463
+008464
+008469
+008471
+008473
+008474
+008476
+008479
+008480
+008481
+008486
+008487
+008488
+008489
+008490
+008491
+008493
+008496
+008497
+008500
+008501
+008504
+008505
+008507
+008508
+008510
+008511
+008515
+008516
+008520
+008525
+008527
+008528
+008531
+008532
+008537
+008538
+008539
+008540
+008543
+008544
+008545
+008546
+008547
+008548
+008551
+008552
+008554
+008555
+008560
+008561
+008563
+008565
+008566
+008567
+008569
+008570
+008571
+008574
+008575
+008577
+008578
+008579
+008580
+008583
+008589
+008590
+008591
+008593
+008594
+008597
+008598
+008599
+008600
+008603
+008605
+008609
+008611
+008613
+008614
+008616
+008619
+008622
+008623
+008625
+008626
+008627
+008629
+008630
+008631
+008632
+008634
+008637
+008640
+008641
+008642
+008643
+008646
+008648
+008649
+008650
+008651
+008652
+008656
+008657
+008658
+008659
+008660
+008661
+008662
+008664
+008666
+008668
+008669
+008671
+008672
+008673
+008674
+008675
+008677
+008678
+008679
+008681
+008682
+008684
+008685
+008686
+008689
+008693
+008694
+008696
+008697
+008700
+008703
+008704
+008705
+008707
+008708
+008711
+008712
+008714
+008715
+008719
+008721
+008724
+008726
+008729
+008734
+008735
+008736
+008737
+008740
+008743
+008745
+008746
+008751
+008754
+008758
+008761
+008762
+008763
+008765
+008767
+008774
+008777
+008778
+008779
+008780
+008781
+008782
+008785
+008786
+008787
+008788
+008789
+008791
+008792
+008795
+008797
+008798
+008800
+008802
+008803
+008804
+008807
+008808
+008812
+008816
+008818
+008820
+008821
+008824
+008825
+008827
+008828
+008829
+008830
+008832
+008834
+008839
+008842
+008844
+008845
+008846
+008850
+008851
+008852
+008853
+008855
+008857
+008860
+008861
+008863
+008864
+008866
+008868
+008869
+008870
+008875
+008877
+008881
+008882
+008887
+008889
+008893
+008894
+008895
+008896
+008897
+008898
+008899
+008901
+008902
+008903
+008904
+008906
+008907
+008908
+008910
+008912
+008915
+008916
+008918
+008922
+008924
+008925
+008928
+008934
+008935
+008937
+008938
+008941
+008945
+008946
+008947
+008949
+008950
+008952
+008954
+008956
+008957
+008959
+008963
+008964
+008972
+008974
+008977
+008981
+008984
+008986
+008990
+008991
+008992
+008993
+008994
+008996
+008998
+009001
+009003
+009008
+009009
+009010
+009011
+009012
+009013
+009014
+009017
+009021
+009023
+009025
+009026
+009028
+009030
+009031
+009033
+009038
+009040
+009041
+009043
+009044
+009046
+009047
+009050
+009052
+009054
+009055
+009056
+009057
+009061
+009062
+009065
+009067
+009069
+009070
+009071
+009074
+009075
+009076
+009077
+009081
+009082
+009083
+009084
+009088
+009090
+009092
+009093
+009095
+009096
+009097
+009101
+009102
+009103
+009104
+009107
+009109
+009110
+009111
+009115
+009118
+009119
+009120
+009122
+009124
+009125
+009127
+009130
+009132
+009134
+009135
+009137
+009139
+009140
+009142
+009143
+009145
+009146
+009149
+009152
+009154
+009156
+009158
+009164
+009165
+009167
+009169
+009170
+009171
+009172
+009176
+009182
+009183
+009188
+009190
+009198
+009199
+009201
+009203
+009204
+009206
+009207
+009210
+009211
+009216
+009217
+009219
+009220
+009222
+009223
+009225
+009226
+009228
+009229
+009231
+009232
+009233
+009234
+009235
+009237
+009240
+009241
+009243
+009248
+009253
+009256
+009257
+009258
+009260
+009261
+009262
+009263
+009264
+009265
+009266
+009267
+009274
+009275
+009276
+009277
+009280
+009284
+009292
+009293
+009294
+009297
+009298
+009300
+009301
+009302
+009304
+009305
+009310
+009311
+009313
+009314
+009317
+009319
+009320
+009321
+009322
+009328
+009329
+009332
+009335
+009338
+009340
+009341
+009344
+009345
+009346
+009352
+009353
+009355
+009356
+009357
+009360
+009361
+009363
+009364
+009366
+009367
+009369
+009370
+009372
+009376
+009379
+009380
+009381
+009383
+009384
+009385
+009387
+009390
+009391
+009395
+009396
+009397
+009399
+009400
+009402
+009403
+009404
+009415
+009416
+009423
+009425
+009426
+009427
+009428
+009430
+009431
+009435
+009436
+009441
+009442
+009444
+009447
+009449
+009450
+009451
+009452
+009453
+009462
+009467
+009471
+009473
+009474
+009475
+009478
+009482
+009483
+009485
+009486
+009487
+009489
+009492
+009493
+009495
+009498
+009501
+009503
+009505
+009506
+009509
+009510
+009511
+009513
+009514
+009521
+009522
+009525
+009529
+009530
+009534
+009535
+009536
+009538
+009539
+009544
+009547
+009548
+009552
+009553
+009554
+009555
+009556
+009559
+009561
+009563
+009564
+009569
+009570
+009572
+009574
+009575
+009578
+009581
+009582
+009583
+009589
+009590
+009592
+009593
+009594
+009595
+009599
+009601
+009602
+009604
+009606
+009607
+009608
+009610
+009612
+009616
+009622
+009624
+009625
+009626
+009628
+009630
+009631
+009632
+009633
+009635
+009639
+009640
+009642
+009643
+009645
+009646
+009648
+009651
+009652
+009653
+009657
+009660
+009661
+009662
+009663
+009665
+009669
+009672
+009673
+009674
+009675
+009677
+009680
+009682
+009683
+009688
+009689
+009690
+009694
+009696
+009697
+009701
+009704
+009705
+009708
+009714
+009715
+009716
+009720
+009722
+009723
+009725
+009727
+009728
+009730
+009731
+009736
+009739
+009740
+009741
+009742
+009744
+009750
+009751
+009752
+009753
+009757
+009759
+009760
+009765
+009766
+009768
+009769
+009770
+009771
+009775
+009777
+009779
+009782
+009783
+009784
+009786
+009787
+009788
+009791
+009793
+009795
+009798
+009799
+009802
+009803
+009804
+009806
+009811
+009812
+009814
+009815
+009817
+009818
+009820
+009821
+009824
+009826
+009827
+009829
+009835
+009837
+009838
+009840
+009843
+009844
+009846
+009847
+009849
+009850
+009853
+009854
+009856
+009857
+009861
+009864
+009866
+009871
+009873
+009875
+009876
+009883
+009885
+009888
+009889
+009890
+009891
+009892
+009893
+009895
+009899
+009901
+009903
+009906
+009907
+009909
+009910
+009912
+009914
+009915
+009916
+009919
+009921
+009922
+009924
+009925
+009927
+009928
+009929
+009930
+009931
+009933
+009934
+009936
+009937
+009941
+009943
+009948
+009951
+009952
+009953
+009956
+009957
+009960
+009962
+009963
diff --git a/ImageSets/Main/train.txt b/ImageSets/Main/train.txt
new file mode 100644
index 0000000..4647eec
--- /dev/null
+++ b/ImageSets/Main/train.txt
@@ -0,0 +1,8218 @@
+000012
+000017
+000023
+000026
+000032
+000033
+000034
+000035
+000036
+000042
+000044
+000047
+000048
+000061
+000064
+000066
+000073
+000077
+000078
+000083
+000089
+000091
+000104
+000112
+000122
+000129
+000133
+000134
+000138
+000140
+000141
+000147
+000153
+000154
+000159
+000161
+000162
+000163
+000164
+000171
+000173
+000174
+000187
+000189
+000192
+000193
+000194
+000198
+000200
+000207
+000209
+000219
+000220
+000222
+000225
+000228
+000235
+000242
+000250
+000256
+000259
+000262
+000263
+000276
+000278
+000282
+000288
+000294
+000296
+000306
+000307
+000311
+000312
+000317
+000320
+000325
+000331
+000334
+000337
+000344
+000347
+000349
+000355
+000359
+000367
+000370
+000372
+000379
+000382
+000387
+000391
+000394
+000395
+000400
+000404
+000406
+000407
+000411
+000416
+000430
+000431
+000438
+000446
+000450
+000454
+000463
+000468
+000469
+000470
+000474
+000476
+000477
+000484
+000489
+000496
+000503
+000508
+000516
+000518
+000519
+000522
+000524
+000525
+000526
+000528
+000535
+000537
+000541
+000544
+000549
+000550
+000552
+000554
+000555
+000559
+000565
+000577
+000583
+000589
+000590
+000592
+000597
+000605
+000609
+000612
+000620
+000622
+000625
+000632
+000633
+000635
+000648
+000654
+000657
+000671
+000672
+000680
+000685
+000688
+000689
+000695
+000699
+000700
+000709
+000710
+000711
+000726
+000729
+000731
+000733
+000739
+000740
+000753
+000754
+000761
+000764
+000767
+000768
+000770
+000774
+000793
+000796
+000804
+000805
+000810
+000812
+000818
+000820
+000822
+000823
+000827
+000828
+000829
+000830
+000845
+000849
+000850
+000851
+000859
+000860
+000865
+000867
+000871
+000887
+000888
+000889
+000892
+000898
+000899
+000900
+000902
+000904
+000906
+000908
+000912
+000915
+000919
+000929
+000936
+000943
+000950
+000951
+000954
+000958
+000962
+000964
+000965
+000966
+000967
+000977
+000980
+000987
+000989
+000991
+000993
+000996
+000997
+000999
+001001
+001002
+001008
+001010
+001011
+001014
+001015
+001024
+001036
+001043
+001050
+001057
+001060
+001064
+001068
+001071
+001073
+001077
+001078
+001079
+001082
+001101
+001106
+001112
+001113
+001119
+001127
+001129
+001130
+001140
+001147
+001151
+001152
+001156
+001158
+001168
+001171
+001172
+001174
+001182
+001191
+001194
+001204
+001205
+001207
+001209
+001212
+001214
+001226
+001229
+001230
+001234
+001237
+001239
+001240
+001248
+001258
+001263
+001268
+001269
+001270
+001273
+001279
+001287
+001294
+001299
+001304
+001309
+001312
+001314
+001315
+001323
+001325
+001326
+001327
+001332
+001333
+001334
+001345
+001346
+001348
+001364
+001365
+001378
+001384
+001385
+001388
+001390
+001393
+001395
+001402
+001404
+001405
+001406
+001408
+001409
+001414
+001418
+001420
+001421
+001426
+001427
+001434
+001436
+001442
+001450
+001451
+001453
+001455
+001457
+001468
+001470
+001479
+001480
+001483
+001485
+001486
+001488
+001492
+001494
+001498
+001499
+001501
+001504
+001512
+001515
+001517
+001521
+001524
+001526
+001528
+001529
+001532
+001539
+001548
+001555
+001556
+001557
+001559
+001563
+001576
+001579
+001580
+001586
+001590
+001593
+001594
+001597
+001604
+001607
+001610
+001611
+001612
+001622
+001627
+001630
+001633
+001636
+001643
+001649
+001650
+001651
+001654
+001661
+001662
+001669
+001673
+001676
+001680
+001683
+001684
+001688
+001690
+001699
+001707
+001708
+001711
+001713
+001714
+001717
+001721
+001723
+001729
+001732
+001733
+001734
+001738
+001739
+001741
+001750
+001752
+001754
+001758
+001759
+001761
+001765
+001766
+001768
+001777
+001780
+001787
+001789
+001800
+001806
+001809
+001810
+001821
+001825
+001828
+001832
+001834
+001836
+001840
+001841
+001843
+001845
+001853
+001854
+001858
+001861
+001864
+001870
+001881
+001892
+001894
+001896
+001898
+001902
+001903
+001904
+001906
+001915
+001922
+001928
+001930
+001937
+001938
+001941
+001945
+001950
+001952
+001954
+001960
+001963
+001971
+001977
+001978
+001980
+001981
+001985
+001989
+001995
+001999
+002001
+002002
+002004
+002006
+002012
+002015
+002020
+002025
+002027
+002034
+002037
+002039
+002042
+002043
+002047
+002049
+002051
+002055
+002056
+002061
+002068
+002069
+002095
+002096
+002104
+002108
+002116
+002117
+002120
+002126
+002132
+002134
+002139
+002151
+002153
+002155
+002156
+002158
+002166
+002170
+002172
+002176
+002178
+002179
+002180
+002182
+002186
+002187
+002191
+002192
+002193
+002194
+002196
+002197
+002199
+002208
+002212
+002215
+002219
+002221
+002224
+002234
+002237
+002238
+002241
+002247
+002249
+002253
+002255
+002256
+002260
+002265
+002277
+002279
+002280
+002284
+002287
+002291
+002293
+002306
+002307
+002310
+002311
+002315
+002318
+002320
+002321
+002323
+002334
+002335
+002342
+002347
+002350
+002354
+002355
+002359
+002362
+002368
+002373
+002384
+002392
+002401
+002403
+002405
+002410
+002411
+002413
+002419
+002420
+002423
+002433
+002436
+002439
+002442
+002443
+002445
+002448
+002458
+002461
+002465
+002466
+002468
+002471
+002472
+002478
+002480
+002481
+002483
+002490
+002494
+002496
+002500
+002501
+002502
+002512
+002514
+002518
+002519
+002533
+002534
+002539
+002544
+002545
+002547
+002554
+002555
+002558
+002559
+002564
+002569
+002571
+002572
+002579
+002590
+002594
+002595
+002599
+002603
+002609
+002611
+002625
+002627
+002634
+002635
+002645
+002647
+002648
+002653
+002662
+002664
+002666
+002669
+002680
+002682
+002683
+002684
+002691
+002697
+002702
+002704
+002710
+002713
+002715
+002722
+002730
+002735
+002737
+002738
+002744
+002745
+002749
+002755
+002757
+002759
+002763
+002765
+002766
+002774
+002778
+002779
+002782
+002783
+002791
+002795
+002796
+002801
+002804
+002807
+002816
+002817
+002820
+002826
+002834
+002841
+002844
+002845
+002848
+002855
+002858
+002864
+002866
+002867
+002868
+002869
+002870
+002873
+002881
+002899
+002906
+002914
+002919
+002931
+002934
+002937
+002939
+002953
+002956
+002957
+002958
+002962
+002969
+002975
+002976
+002987
+002988
+002989
+002990
+002992
+002995
+003002
+003003
+003007
+003011
+003013
+003024
+003027
+003034
+003042
+003047
+003051
+003053
+003061
+003063
+003066
+003074
+003077
+003083
+003085
+003088
+003092
+003100
+003103
+003105
+003106
+003107
+003108
+003110
+003116
+003122
+003124
+003133
+003134
+003135
+003138
+003140
+003145
+003146
+003147
+003149
+003150
+003155
+003157
+003159
+003161
+003163
+003165
+003169
+003175
+003181
+003183
+003184
+003185
+003188
+003202
+003204
+003205
+003211
+003214
+003229
+003231
+003233
+003236
+003240
+003242
+003244
+003247
+003253
+003254
+003259
+003260
+003261
+003269
+003270
+003273
+003279
+003280
+003282
+003284
+003290
+003292
+003303
+003308
+003320
+003330
+003331
+003336
+003337
+003338
+003339
+003343
+003349
+003350
+003354
+003355
+003356
+003359
+003363
+003365
+003367
+003369
+003373
+003374
+003379
+003380
+003382
+003392
+003395
+003396
+003401
+003406
+003408
+003412
+003413
+003416
+003417
+003420
+003421
+003424
+003430
+003433
+003436
+003439
+003441
+003450
+003452
+003466
+003477
+003484
+003487
+003489
+003491
+003493
+003496
+003497
+003499
+003500
+003506
+003508
+003509
+003510
+003511
+003522
+003524
+003525
+003529
+003539
+003548
+003549
+003550
+003551
+003555
+003564
+003565
+003575
+003576
+003577
+003585
+003586
+003588
+003596
+003599
+003603
+003604
+003605
+003608
+003609
+003614
+003621
+003622
+003625
+003627
+003628
+003629
+003634
+003635
+003642
+003644
+003645
+003646
+003656
+003658
+003662
+003663
+003664
+003671
+003678
+003679
+003681
+003688
+003694
+003695
+003698
+003699
+003700
+003704
+003705
+003713
+003714
+003732
+003735
+003740
+003743
+003748
+003749
+003751
+003752
+003758
+003759
+003763
+003767
+003773
+003779
+003781
+003784
+003786
+003788
+003790
+003792
+003797
+003806
+003807
+003811
+003817
+003818
+003824
+003827
+003828
+003830
+003834
+003835
+003847
+003849
+003856
+003859
+003860
+003861
+003865
+003866
+003874
+003879
+003887
+003889
+003890
+003898
+003899
+003907
+003912
+003913
+003921
+003932
+003935
+003936
+003939
+003945
+003949
+003953
+003956
+003961
+003969
+003970
+003971
+003974
+003983
+003987
+003988
+003991
+003993
+003997
+003998
+004005
+004008
+004009
+004012
+004013
+004014
+004016
+004017
+004019
+004023
+004028
+004033
+004034
+004035
+004037
+004046
+004052
+004058
+004067
+004091
+004092
+004093
+004095
+004100
+004106
+004111
+004120
+004121
+004129
+004131
+004133
+004136
+004137
+004138
+004140
+004146
+004149
+004152
+004158
+004163
+004164
+004168
+004169
+004170
+004171
+004189
+004190
+004196
+004200
+004201
+004209
+004215
+004220
+004221
+004223
+004224
+004228
+004231
+004232
+004237
+004241
+004242
+004244
+004247
+004253
+004255
+004256
+004263
+004269
+004270
+004271
+004272
+004273
+004280
+004281
+004283
+004287
+004291
+004292
+004296
+004300
+004303
+004307
+004315
+004318
+004322
+004325
+004327
+004333
+004338
+004339
+004345
+004347
+004359
+004360
+004361
+004365
+004367
+004370
+004371
+004372
+004376
+004379
+004386
+004387
+004389
+004391
+004392
+004404
+004434
+004436
+004439
+004441
+004452
+004470
+004471
+004479
+004481
+004484
+004496
+004500
+004502
+004508
+004510
+004514
+004517
+004519
+004520
+004524
+004526
+004537
+004540
+004544
+004548
+004549
+004551
+004553
+004562
+004563
+004565
+004566
+004570
+004571
+004576
+004579
+004584
+004587
+004591
+004595
+004597
+004604
+004605
+004607
+004611
+004612
+004622
+004623
+004625
+004627
+004628
+004631
+004634
+004636
+004643
+004644
+004648
+004651
+004656
+004671
+004675
+004679
+004683
+004685
+004686
+004687
+004691
+004693
+004694
+004701
+004702
+004705
+004706
+004710
+004714
+004715
+004718
+004723
+004735
+004737
+004742
+004743
+004747
+004748
+004753
+004754
+004760
+004773
+004776
+004779
+004782
+004783
+004790
+004792
+004793
+004794
+004797
+004799
+004801
+004808
+004815
+004823
+004828
+004830
+004832
+004836
+004837
+004841
+004842
+004846
+004848
+004849
+004857
+004869
+004873
+004876
+004879
+004882
+004885
+004897
+004898
+004902
+004905
+004907
+004910
+004911
+004913
+004929
+004946
+004951
+004955
+004958
+004961
+004962
+004966
+004968
+004972
+004973
+004974
+004976
+004984
+004987
+004990
+004991
+004992
+004995
+005001
+005004
+005006
+005007
+005016
+005018
+005020
+005023
+005024
+005026
+005027
+005029
+005032
+005033
+005045
+005047
+005052
+005057
+005058
+005061
+005065
+005068
+005071
+005073
+005078
+005084
+005086
+005090
+005093
+005094
+005097
+005101
+005107
+005108
+005114
+005121
+005122
+005124
+005129
+005130
+005134
+005138
+005143
+005153
+005156
+005168
+005169
+005171
+005173
+005177
+005181
+005183
+005186
+005189
+005190
+005191
+005202
+005203
+005208
+005215
+005217
+005219
+005223
+005231
+005236
+005244
+005245
+005246
+005257
+005258
+005259
+005260
+005262
+005269
+005273
+005283
+005285
+005288
+005290
+005292
+005297
+005303
+005304
+005307
+005310
+005311
+005318
+005327
+005336
+005337
+005338
+005344
+005345
+005351
+005358
+005360
+005363
+005368
+005369
+005373
+005374
+005387
+005388
+005389
+005391
+005396
+005404
+005405
+005406
+005408
+005410
+005413
+005414
+005417
+005420
+005424
+005433
+005440
+005445
+005448
+005450
+005451
+005453
+005455
+005457
+005467
+005478
+005483
+005487
+005489
+005496
+005499
+005508
+005509
+005511
+005514
+005515
+005519
+005524
+005526
+005527
+005536
+005541
+005542
+005544
+005547
+005563
+005566
+005568
+005574
+005579
+005582
+005585
+005591
+005592
+005599
+005600
+005601
+005603
+005605
+005609
+005611
+005624
+005625
+005630
+005631
+005636
+005637
+005639
+005644
+005648
+005654
+005658
+005668
+005669
+005680
+005686
+005695
+005697
+005699
+005700
+005704
+005705
+005710
+005713
+005715
+005718
+005728
+005730
+005731
+005735
+005738
+005740
+005742
+005752
+005756
+005757
+005764
+005765
+005769
+005780
+005782
+005783
+005784
+005786
+005789
+005796
+005803
+005805
+005806
+005813
+005814
+005817
+005821
+005824
+005826
+005831
+005836
+005838
+005840
+005843
+005850
+005851
+005859
+005860
+005861
+005864
+005867
+005873
+005881
+005884
+005885
+005888
+005889
+005893
+005895
+005899
+005901
+005903
+005905
+005908
+005909
+005910
+005911
+005918
+005920
+005923
+005930
+005938
+005947
+005948
+005951
+005960
+005961
+005964
+005971
+005980
+005983
+005984
+005990
+005992
+006004
+006009
+006011
+006020
+006023
+006025
+006030
+006033
+006038
+006043
+006061
+006065
+006066
+006067
+006070
+006073
+006074
+006078
+006079
+006088
+006091
+006095
+006096
+006100
+006103
+006104
+006105
+006123
+006128
+006130
+006131
+006134
+006135
+006140
+006141
+006156
+006158
+006162
+006166
+006170
+006171
+006172
+006174
+006175
+006176
+006177
+006179
+006180
+006181
+006183
+006187
+006189
+006196
+006208
+006210
+006221
+006223
+006224
+006225
+006229
+006230
+006236
+006238
+006243
+006247
+006250
+006251
+006261
+006262
+006264
+006267
+006270
+006272
+006275
+006279
+006285
+006289
+006290
+006291
+006299
+006304
+006305
+006320
+006329
+006341
+006344
+006349
+006352
+006353
+006362
+006363
+006366
+006367
+006369
+006371
+006374
+006375
+006381
+006382
+006395
+006400
+006411
+006417
+006418
+006419
+006427
+006429
+006433
+006434
+006436
+006438
+006442
+006447
+006448
+006455
+006458
+006459
+006462
+006466
+006470
+006472
+006474
+006475
+006476
+006482
+006483
+006486
+006495
+006499
+006501
+006503
+006506
+006515
+006523
+006524
+006536
+006547
+006548
+006549
+006550
+006551
+006556
+006560
+006564
+006569
+006595
+006597
+006602
+006605
+006609
+006610
+006612
+006622
+006626
+006627
+006635
+006636
+006637
+006638
+006648
+006652
+006654
+006658
+006660
+006674
+006684
+006689
+006694
+006695
+006697
+006698
+006703
+006704
+006706
+006707
+006708
+006714
+006726
+006727
+006731
+006734
+006735
+006736
+006738
+006740
+006748
+006753
+006755
+006766
+006773
+006777
+006781
+006782
+006784
+006794
+006805
+006806
+006810
+006822
+006824
+006825
+006833
+006836
+006839
+006840
+006844
+006845
+006847
+006848
+006849
+006852
+006858
+006864
+006866
+006868
+006869
+006874
+006883
+006887
+006893
+006896
+006899
+006900
+006909
+006910
+006911
+006912
+006914
+006916
+006917
+006919
+006930
+006931
+006939
+006943
+006947
+006948
+006950
+006958
+006959
+006968
+006971
+006976
+006983
+007002
+007003
+007006
+007007
+007011
+007016
+007018
+007023
+007025
+007029
+007033
+007036
+007039
+007040
+007045
+007050
+007062
+007064
+007072
+007073
+007075
+007078
+007079
+007080
+007088
+007089
+007090
+007092
+007093
+007095
+007105
+007108
+007113
+007121
+007125
+007128
+007129
+007130
+007133
+007138
+007150
+007152
+007154
+007159
+007163
+007166
+007168
+007177
+007180
+007182
+007184
+007185
+007193
+007194
+007197
+007205
+007213
+007214
+007219
+007222
+007223
+007234
+007241
+007243
+007250
+007256
+007261
+007263
+007271
+007279
+007285
+007289
+007295
+007298
+007305
+007308
+007322
+007323
+007325
+007327
+007334
+007336
+007351
+007361
+007365
+007369
+007370
+007373
+007375
+007381
+007385
+007389
+007394
+007396
+007398
+007410
+007411
+007413
+007417
+007419
+007421
+007425
+007431
+007437
+007446
+007454
+007458
+007466
+007467
+007468
+007474
+007477
+007479
+007481
+007483
+007490
+007491
+007493
+007497
+007503
+007513
+007519
+007521
+007524
+007526
+007530
+007535
+007536
+007538
+007540
+007544
+007558
+007565
+007566
+007570
+007572
+007575
+007578
+007586
+007590
+007594
+007600
+007601
+007606
+007611
+007619
+007621
+007629
+007631
+007633
+007637
+007653
+007654
+007655
+007663
+007667
+007683
+007685
+007692
+007696
+007697
+007699
+007704
+007713
+007718
+007721
+007729
+007731
+007735
+007736
+007740
+007748
+007749
+007751
+007753
+007762
+007767
+007775
+007777
+007781
+007790
+007791
+007795
+007803
+007809
+007810
+007814
+007819
+007820
+007821
+007831
+007836
+007838
+007840
+007847
+007853
+007854
+007859
+007863
+007864
+007872
+007876
+007877
+007878
+007883
+007884
+007885
+007898
+007900
+007901
+007905
+007908
+007910
+007911
+007914
+007915
+007923
+007925
+007926
+007932
+007939
+007940
+007953
+007959
+007963
+007964
+007968
+007974
+007976
+007980
+007991
+007996
+008001
+008004
+008005
+008008
+008012
+008017
+008019
+008026
+008037
+008040
+008042
+008043
+008044
+008049
+008051
+008053
+008062
+008063
+008064
+008067
+008072
+008075
+008076
+008079
+008082
+008083
+008084
+008093
+008095
+008096
+008098
+008106
+008108
+008116
+008117
+008121
+008127
+008130
+008137
+008139
+008142
+008150
+008163
+008164
+008166
+008169
+008174
+008186
+008188
+008197
+008199
+008202
+008203
+008204
+008211
+008213
+008216
+008218
+008223
+008226
+008232
+008235
+008248
+008250
+008252
+008253
+008254
+008260
+008261
+008262
+008263
+008269
+008272
+008280
+008282
+008296
+008301
+008302
+008310
+008311
+008312
+008313
+008315
+008316
+008317
+008322
+008332
+008336
+008338
+008341
+008342
+008346
+008351
+008360
+008372
+008374
+008381
+008384
+008385
+008388
+008391
+008397
+008398
+008403
+008409
+008422
+008425
+008426
+008427
+008437
+008442
+008443
+008445
+008449
+008452
+008453
+008456
+008462
+008465
+008466
+008467
+008468
+008470
+008475
+008477
+008478
+008482
+008483
+008495
+008506
+008517
+008523
+008529
+008530
+008533
+008536
+008549
+008550
+008558
+008559
+008568
+008581
+008585
+008587
+008588
+008595
+008596
+008602
+008610
+008615
+008617
+008618
+008628
+008633
+008645
+008655
+008663
+008665
+008670
+008676
+008688
+008690
+008691
+008699
+008702
+008706
+008710
+008720
+008723
+008725
+008727
+008731
+008732
+008738
+008741
+008744
+008748
+008750
+008755
+008756
+008757
+008760
+008764
+008768
+008770
+008771
+008776
+008783
+008784
+008790
+008794
+008806
+008809
+008811
+008813
+008814
+008815
+008819
+008838
+008840
+008841
+008847
+008856
+008862
+008865
+008872
+008878
+008879
+008883
+008885
+008886
+008891
+008900
+008905
+008909
+008920
+008923
+008926
+008929
+008930
+008932
+008933
+008936
+008939
+008944
+008948
+008958
+008960
+008961
+008962
+008966
+008967
+008968
+008969
+008970
+008971
+008973
+008975
+008978
+008979
+008980
+008985
+008987
+008988
+008989
+008995
+008999
+009000
+009004
+009005
+009016
+009018
+009020
+009027
+009029
+009032
+009036
+009042
+009045
+009049
+009058
+009059
+009063
+009066
+009068
+009073
+009078
+009080
+009086
+009098
+009099
+009100
+009106
+009108
+009114
+009117
+009121
+009123
+009136
+009144
+009148
+009153
+009160
+009161
+009166
+009173
+009175
+009181
+009184
+009185
+009191
+009196
+009197
+009200
+009205
+009208
+009209
+009214
+009215
+009218
+009227
+009230
+009238
+009242
+009245
+009251
+009252
+009255
+009259
+009269
+009270
+009271
+009272
+009283
+009285
+009287
+009288
+009289
+009290
+009295
+009296
+009299
+009306
+009307
+009308
+009316
+009318
+009324
+009325
+009327
+009333
+009336
+009339
+009342
+009343
+009358
+009359
+009362
+009365
+009377
+009386
+009388
+009389
+009392
+009393
+009394
+009398
+009406
+009407
+009409
+009410
+009411
+009413
+009417
+009418
+009419
+009420
+009421
+009422
+009424
+009429
+009432
+009434
+009446
+009458
+009460
+009463
+009465
+009466
+009469
+009476
+009488
+009490
+009491
+009496
+009497
+009499
+009504
+009508
+009512
+009515
+009516
+009518
+009520
+009523
+009524
+009526
+009528
+009537
+009541
+009542
+009545
+009549
+009551
+009557
+009562
+009566
+009573
+009576
+009577
+009579
+009584
+009585
+009587
+009596
+009600
+009605
+009609
+009613
+009614
+009615
+009618
+009621
+009623
+009629
+009634
+009637
+009638
+009644
+009650
+009654
+009656
+009659
+009664
+009666
+009668
+009671
+009679
+009684
+009691
+009693
+009702
+009703
+009707
+009709
+009713
+009717
+009718
+009721
+009729
+009733
+009734
+009735
+009749
+009755
+009756
+009762
+009763
+009774
+009776
+009789
+009790
+009792
+009797
+009800
+009805
+009807
+009808
+009810
+009813
+009825
+009828
+009830
+009832
+009834
+009839
+009842
+009845
+009848
+009851
+009852
+009855
+009859
+009860
+009867
+009868
+009869
+009872
+009874
+009877
+009878
+009879
+009882
+009884
+009887
+009896
+009904
+009911
+009918
+009920
+009926
+009938
+009940
+009942
+009944
+009945
+009949
+009959
+009961
+2008_000008
+2008_000015
+2008_000019
+2008_000023
+2008_000028
+2008_000033
+2008_000036
+2008_000037
+2008_000041
+2008_000045
+2008_000053
+2008_000060
+2008_000066
+2008_000070
+2008_000074
+2008_000085
+2008_000089
+2008_000093
+2008_000095
+2008_000096
+2008_000097
+2008_000099
+2008_000103
+2008_000105
+2008_000109
+2008_000112
+2008_000128
+2008_000131
+2008_000132
+2008_000141
+2008_000142
+2008_000143
+2008_000144
+2008_000148
+2008_000151
+2008_000154
+2008_000162
+2008_000176
+2008_000181
+2008_000185
+2008_000187
+2008_000188
+2008_000189
+2008_000191
+2008_000192
+2008_000193
+2008_000196
+2008_000197
+2008_000199
+2008_000202
+2008_000207
+2008_000217
+2008_000226
+2008_000227
+2008_000235
+2008_000236
+2008_000237
+2008_000238
+2008_000252
+2008_000255
+2008_000259
+2008_000260
+2008_000262
+2008_000266
+2008_000273
+2008_000275
+2008_000283
+2008_000284
+2008_000287
+2008_000289
+2008_000290
+2008_000291
+2008_000297
+2008_000309
+2008_000311
+2008_000313
+2008_000315
+2008_000316
+2008_000318
+2008_000330
+2008_000335
+2008_000336
+2008_000338
+2008_000342
+2008_000343
+2008_000346
+2008_000348
+2008_000350
+2008_000356
+2008_000361
+2008_000364
+2008_000365
+2008_000371
+2008_000380
+2008_000392
+2008_000393
+2008_000397
+2008_000399
+2008_000400
+2008_000405
+2008_000415
+2008_000416
+2008_000421
+2008_000422
+2008_000426
+2008_000428
+2008_000432
+2008_000435
+2008_000436
+2008_000437
+2008_000442
+2008_000443
+2008_000445
+2008_000447
+2008_000448
+2008_000455
+2008_000461
+2008_000470
+2008_000471
+2008_000480
+2008_000488
+2008_000491
+2008_000493
+2008_000495
+2008_000499
+2008_000502
+2008_000505
+2008_000512
+2008_000514
+2008_000515
+2008_000527
+2008_000531
+2008_000540
+2008_000544
+2008_000545
+2008_000548
+2008_000552
+2008_000559
+2008_000561
+2008_000563
+2008_000567
+2008_000572
+2008_000578
+2008_000583
+2008_000584
+2008_000585
+2008_000588
+2008_000595
+2008_000607
+2008_000613
+2008_000615
+2008_000619
+2008_000626
+2008_000628
+2008_000636
+2008_000641
+2008_000645
+2008_000646
+2008_000648
+2008_000650
+2008_000655
+2008_000669
+2008_000672
+2008_000674
+2008_000676
+2008_000678
+2008_000683
+2008_000689
+2008_000694
+2008_000696
+2008_000703
+2008_000704
+2008_000711
+2008_000716
+2008_000719
+2008_000721
+2008_000723
+2008_000724
+2008_000726
+2008_000729
+2008_000732
+2008_000733
+2008_000742
+2008_000753
+2008_000756
+2008_000758
+2008_000760
+2008_000761
+2008_000764
+2008_000775
+2008_000777
+2008_000778
+2008_000785
+2008_000787
+2008_000790
+2008_000792
+2008_000798
+2008_000801
+2008_000808
+2008_000814
+2008_000815
+2008_000824
+2008_000829
+2008_000832
+2008_000833
+2008_000841
+2008_000842
+2008_000844
+2008_000847
+2008_000851
+2008_000854
+2008_000860
+2008_000861
+2008_000867
+2008_000870
+2008_000873
+2008_000875
+2008_000881
+2008_000883
+2008_000887
+2008_000899
+2008_000901
+2008_000902
+2008_000905
+2008_000908
+2008_000912
+2008_000914
+2008_000915
+2008_000923
+2008_000924
+2008_000928
+2008_000934
+2008_000941
+2008_000944
+2008_000953
+2008_000959
+2008_000970
+2008_000973
+2008_000979
+2008_000981
+2008_000985
+2008_000987
+2008_000999
+2008_001018
+2008_001020
+2008_001021
+2008_001022
+2008_001023
+2008_001026
+2008_001030
+2008_001031
+2008_001035
+2008_001036
+2008_001039
+2008_001042
+2008_001047
+2008_001048
+2008_001052
+2008_001054
+2008_001056
+2008_001057
+2008_001071
+2008_001073
+2008_001081
+2008_001083
+2008_001104
+2008_001105
+2008_001106
+2008_001112
+2008_001115
+2008_001118
+2008_001119
+2008_001130
+2008_001133
+2008_001134
+2008_001137
+2008_001143
+2008_001147
+2008_001158
+2008_001159
+2008_001161
+2008_001164
+2008_001169
+2008_001171
+2008_001182
+2008_001188
+2008_001189
+2008_001190
+2008_001196
+2008_001202
+2008_001203
+2008_001206
+2008_001208
+2008_001215
+2008_001219
+2008_001223
+2008_001230
+2008_001235
+2008_001238
+2008_001245
+2008_001263
+2008_001267
+2008_001272
+2008_001274
+2008_001278
+2008_001285
+2008_001294
+2008_001299
+2008_001302
+2008_001307
+2008_001310
+2008_001312
+2008_001325
+2008_001329
+2008_001335
+2008_001336
+2008_001346
+2008_001351
+2008_001357
+2008_001358
+2008_001359
+2008_001373
+2008_001375
+2008_001382
+2008_001383
+2008_001385
+2008_001387
+2008_001389
+2008_001390
+2008_001399
+2008_001402
+2008_001405
+2008_001408
+2008_001413
+2008_001414
+2008_001419
+2008_001420
+2008_001431
+2008_001434
+2008_001440
+2008_001444
+2008_001446
+2008_001448
+2008_001454
+2008_001455
+2008_001460
+2008_001461
+2008_001462
+2008_001464
+2008_001467
+2008_001479
+2008_001482
+2008_001488
+2008_001493
+2008_001495
+2008_001498
+2008_001500
+2008_001501
+2008_001510
+2008_001523
+2008_001525
+2008_001529
+2008_001533
+2008_001538
+2008_001541
+2008_001550
+2008_001563
+2008_001566
+2008_001576
+2008_001577
+2008_001582
+2008_001591
+2008_001592
+2008_001601
+2008_001609
+2008_001610
+2008_001615
+2008_001617
+2008_001620
+2008_001626
+2008_001631
+2008_001632
+2008_001641
+2008_001643
+2008_001645
+2008_001652
+2008_001653
+2008_001661
+2008_001666
+2008_001670
+2008_001673
+2008_001679
+2008_001690
+2008_001691
+2008_001692
+2008_001694
+2008_001699
+2008_001704
+2008_001706
+2008_001708
+2008_001709
+2008_001710
+2008_001716
+2008_001719
+2008_001724
+2008_001729
+2008_001735
+2008_001737
+2008_001741
+2008_001744
+2008_001746
+2008_001751
+2008_001758
+2008_001761
+2008_001770
+2008_001775
+2008_001781
+2008_001783
+2008_001787
+2008_001789
+2008_001791
+2008_001796
+2008_001797
+2008_001801
+2008_001809
+2008_001811
+2008_001813
+2008_001829
+2008_001832
+2008_001834
+2008_001836
+2008_001837
+2008_001842
+2008_001845
+2008_001849
+2008_001852
+2008_001854
+2008_001856
+2008_001860
+2008_001865
+2008_001866
+2008_001872
+2008_001876
+2008_001880
+2008_001881
+2008_001882
+2008_001888
+2008_001894
+2008_001896
+2008_001903
+2008_001911
+2008_001921
+2008_001926
+2008_001929
+2008_001937
+2008_001941
+2008_001947
+2008_001955
+2008_001956
+2008_001957
+2008_001967
+2008_001970
+2008_001977
+2008_001980
+2008_001982
+2008_001986
+2008_001997
+2008_002000
+2008_002001
+2008_002002
+2008_002005
+2008_002009
+2008_002023
+2008_002032
+2008_002033
+2008_002056
+2008_002058
+2008_002061
+2008_002062
+2008_002064
+2008_002066
+2008_002067
+2008_002073
+2008_002079
+2008_002080
+2008_002093
+2008_002094
+2008_002096
+2008_002103
+2008_002112
+2008_002116
+2008_002117
+2008_002118
+2008_002119
+2008_002123
+2008_002129
+2008_002131
+2008_002145
+2008_002148
+2008_002150
+2008_002156
+2008_002160
+2008_002162
+2008_002175
+2008_002177
+2008_002181
+2008_002182
+2008_002195
+2008_002197
+2008_002200
+2008_002202
+2008_002204
+2008_002206
+2008_002208
+2008_002210
+2008_002215
+2008_002218
+2008_002220
+2008_002221
+2008_002225
+2008_002227
+2008_002229
+2008_002236
+2008_002243
+2008_002244
+2008_002247
+2008_002248
+2008_002251
+2008_002255
+2008_002258
+2008_002262
+2008_002270
+2008_002278
+2008_002279
+2008_002280
+2008_002281
+2008_002288
+2008_002294
+2008_002296
+2008_002299
+2008_002304
+2008_002307
+2008_002311
+2008_002317
+2008_002325
+2008_002327
+2008_002329
+2008_002331
+2008_002335
+2008_002338
+2008_002340
+2008_002343
+2008_002344
+2008_002350
+2008_002357
+2008_002361
+2008_002362
+2008_002365
+2008_002368
+2008_002369
+2008_002370
+2008_002377
+2008_002389
+2008_002399
+2008_002405
+2008_002411
+2008_002418
+2008_002422
+2008_002425
+2008_002434
+2008_002437
+2008_002441
+2008_002442
+2008_002448
+2008_002457
+2008_002458
+2008_002459
+2008_002461
+2008_002465
+2008_002466
+2008_002471
+2008_002473
+2008_002482
+2008_002484
+2008_002487
+2008_002491
+2008_002501
+2008_002506
+2008_002514
+2008_002515
+2008_002524
+2008_002533
+2008_002541
+2008_002543
+2008_002547
+2008_002549
+2008_002551
+2008_002555
+2008_002562
+2008_002566
+2008_002568
+2008_002574
+2008_002575
+2008_002578
+2008_002583
+2008_002584
+2008_002601
+2008_002612
+2008_002613
+2008_002621
+2008_002622
+2008_002625
+2008_002634
+2008_002638
+2008_002641
+2008_002647
+2008_002648
+2008_002649
+2008_002650
+2008_002662
+2008_002665
+2008_002666
+2008_002668
+2008_002672
+2008_002674
+2008_002675
+2008_002676
+2008_002679
+2008_002686
+2008_002697
+2008_002698
+2008_002704
+2008_002710
+2008_002712
+2008_002718
+2008_002719
+2008_002728
+2008_002730
+2008_002733
+2008_002736
+2008_002741
+2008_002749
+2008_002750
+2008_002758
+2008_002760
+2008_002762
+2008_002767
+2008_002772
+2008_002774
+2008_002776
+2008_002784
+2008_002787
+2008_002791
+2008_002793
+2008_002794
+2008_002804
+2008_002808
+2008_002813
+2008_002823
+2008_002834
+2008_002842
+2008_002848
+2008_002850
+2008_002854
+2008_002856
+2008_002857
+2008_002866
+2008_002868
+2008_002872
+2008_002873
+2008_002880
+2008_002885
+2008_002887
+2008_002890
+2008_002891
+2008_002892
+2008_002894
+2008_002903
+2008_002913
+2008_002916
+2008_002917
+2008_002922
+2008_002930
+2008_002931
+2008_002943
+2008_002948
+2008_002951
+2008_002954
+2008_002955
+2008_002957
+2008_002960
+2008_002961
+2008_002966
+2008_002970
+2008_002972
+2008_002977
+2008_002983
+2008_002984
+2008_002985
+2008_002988
+2008_002993
+2008_002997
+2008_003013
+2008_003015
+2008_003017
+2008_003018
+2008_003021
+2008_003023
+2008_003025
+2008_003033
+2008_003037
+2008_003039
+2008_003041
+2008_003043
+2008_003048
+2008_003049
+2008_003057
+2008_003059
+2008_003060
+2008_003061
+2008_003063
+2008_003065
+2008_003068
+2008_003075
+2008_003079
+2008_003081
+2008_003083
+2008_003087
+2008_003093
+2008_003094
+2008_003099
+2008_003101
+2008_003112
+2008_003114
+2008_003120
+2008_003122
+2008_003127
+2008_003128
+2008_003134
+2008_003140
+2008_003143
+2008_003146
+2008_003147
+2008_003151
+2008_003154
+2008_003157
+2008_003160
+2008_003168
+2008_003180
+2008_003182
+2008_003191
+2008_003196
+2008_003200
+2008_003203
+2008_003208
+2008_003209
+2008_003213
+2008_003224
+2008_003231
+2008_003242
+2008_003244
+2008_003248
+2008_003251
+2008_003252
+2008_003261
+2008_003264
+2008_003265
+2008_003266
+2008_003269
+2008_003272
+2008_003275
+2008_003276
+2008_003277
+2008_003283
+2008_003287
+2008_003288
+2008_003290
+2008_003297
+2008_003302
+2008_003303
+2008_003304
+2008_003311
+2008_003313
+2008_003318
+2008_003321
+2008_003323
+2008_003329
+2008_003335
+2008_003338
+2008_003342
+2008_003347
+2008_003360
+2008_003362
+2008_003373
+2008_003378
+2008_003380
+2008_003381
+2008_003386
+2008_003393
+2008_003394
+2008_003406
+2008_003409
+2008_003414
+2008_003415
+2008_003417
+2008_003418
+2008_003426
+2008_003429
+2008_003430
+2008_003434
+2008_003435
+2008_003437
+2008_003447
+2008_003448
+2008_003452
+2008_003458
+2008_003462
+2008_003463
+2008_003469
+2008_003478
+2008_003480
+2008_003485
+2008_003488
+2008_003489
+2008_003496
+2008_003497
+2008_003498
+2008_003500
+2008_003501
+2008_003504
+2008_003507
+2008_003510
+2008_003515
+2008_003520
+2008_003521
+2008_003522
+2008_003523
+2008_003533
+2008_003534
+2008_003544
+2008_003559
+2008_003560
+2008_003562
+2008_003571
+2008_003575
+2008_003578
+2008_003579
+2008_003582
+2008_003585
+2008_003587
+2008_003589
+2008_003590
+2008_003596
+2008_003608
+2008_003611
+2008_003617
+2008_003622
+2008_003626
+2008_003629
+2008_003635
+2008_003637
+2008_003645
+2008_003652
+2008_003653
+2008_003655
+2008_003659
+2008_003665
+2008_003667
+2008_003674
+2008_003675
+2008_003677
+2008_003682
+2008_003685
+2008_003688
+2008_003689
+2008_003691
+2008_003697
+2008_003701
+2008_003703
+2008_003706
+2008_003707
+2008_003712
+2008_003719
+2008_003726
+2008_003729
+2008_003732
+2008_003746
+2008_003748
+2008_003761
+2008_003762
+2008_003764
+2008_003769
+2008_003773
+2008_003774
+2008_003776
+2008_003779
+2008_003781
+2008_003788
+2008_003791
+2008_003796
+2008_003802
+2008_003811
+2008_003814
+2008_003815
+2008_003819
+2008_003831
+2008_003841
+2008_003842
+2008_003847
+2008_003849
+2008_003852
+2008_003854
+2008_003864
+2008_003866
+2008_003870
+2008_003871
+2008_003882
+2008_003883
+2008_003888
+2008_003891
+2008_003892
+2008_003908
+2008_003913
+2008_003914
+2008_003916
+2008_003920
+2008_003922
+2008_003925
+2008_003939
+2008_003942
+2008_003947
+2008_003956
+2008_003966
+2008_003967
+2008_003970
+2008_003974
+2008_003975
+2008_003978
+2008_003984
+2008_003985
+2008_003986
+2008_003992
+2008_003995
+2008_003998
+2008_004000
+2008_004004
+2008_004008
+2008_004014
+2008_004017
+2008_004021
+2008_004022
+2008_004024
+2008_004026
+2008_004036
+2008_004037
+2008_004042
+2008_004044
+2008_004053
+2008_004055
+2008_004066
+2008_004074
+2008_004077
+2008_004080
+2008_004084
+2008_004087
+2008_004092
+2008_004097
+2008_004100
+2008_004102
+2008_004106
+2008_004110
+2008_004112
+2008_004113
+2008_004120
+2008_004122
+2008_004130
+2008_004134
+2008_004138
+2008_004145
+2008_004147
+2008_004148
+2008_004161
+2008_004163
+2008_004165
+2008_004171
+2008_004176
+2008_004195
+2008_004196
+2008_004201
+2008_004208
+2008_004217
+2008_004218
+2008_004224
+2008_004231
+2008_004232
+2008_004235
+2008_004239
+2008_004246
+2008_004247
+2008_004259
+2008_004265
+2008_004269
+2008_004274
+2008_004276
+2008_004280
+2008_004284
+2008_004287
+2008_004288
+2008_004291
+2008_004293
+2008_004296
+2008_004301
+2008_004303
+2008_004307
+2008_004313
+2008_004314
+2008_004319
+2008_004321
+2008_004325
+2008_004328
+2008_004331
+2008_004342
+2008_004353
+2008_004358
+2008_004362
+2008_004365
+2008_004372
+2008_004376
+2008_004378
+2008_004380
+2008_004385
+2008_004387
+2008_004398
+2008_004403
+2008_004410
+2008_004411
+2008_004412
+2008_004416
+2008_004418
+2008_004428
+2008_004430
+2008_004435
+2008_004436
+2008_004439
+2008_004441
+2008_004443
+2008_004450
+2008_004452
+2008_004457
+2008_004458
+2008_004462
+2008_004480
+2008_004488
+2008_004490
+2008_004492
+2008_004493
+2008_004499
+2008_004505
+2008_004506
+2008_004512
+2008_004513
+2008_004515
+2008_004518
+2008_004519
+2008_004532
+2008_004539
+2008_004544
+2008_004545
+2008_004547
+2008_004551
+2008_004559
+2008_004567
+2008_004568
+2008_004570
+2008_004574
+2008_004581
+2008_004583
+2008_004584
+2008_004585
+2008_004588
+2008_004590
+2008_004593
+2008_004602
+2008_004603
+2008_004607
+2008_004611
+2008_004616
+2008_004617
+2008_004620
+2008_004629
+2008_004631
+2008_004634
+2008_004635
+2008_004636
+2008_004648
+2008_004649
+2008_004661
+2008_004663
+2008_004666
+2008_004667
+2008_004668
+2008_004671
+2008_004672
+2008_004677
+2008_004678
+2008_004679
+2008_004690
+2008_004692
+2008_004697
+2008_004703
+2008_004707
+2008_004713
+2008_004719
+2008_004725
+2008_004732
+2008_004739
+2008_004749
+2008_004750
+2008_004752
+2008_004763
+2008_004764
+2008_004767
+2008_004770
+2008_004776
+2008_004777
+2008_004781
+2008_004783
+2008_004786
+2008_004802
+2008_004804
+2008_004807
+2008_004808
+2008_004821
+2008_004822
+2008_004827
+2008_004834
+2008_004838
+2008_004841
+2008_004844
+2008_004845
+2008_004847
+2008_004849
+2008_004850
+2008_004856
+2008_004858
+2008_004868
+2008_004869
+2008_004872
+2008_004874
+2008_004876
+2008_004892
+2008_004893
+2008_004899
+2008_004903
+2008_004908
+2008_004911
+2008_004914
+2008_004917
+2008_004920
+2008_004931
+2008_004934
+2008_004937
+2008_004938
+2008_004945
+2008_004946
+2008_004950
+2008_004961
+2008_004964
+2008_004966
+2008_004969
+2008_004970
+2008_004973
+2008_004976
+2008_004977
+2008_004981
+2008_004983
+2008_004985
+2008_004990
+2008_004991
+2008_004998
+2008_005000
+2008_005006
+2008_005013
+2008_005016
+2008_005033
+2008_005036
+2008_005040
+2008_005042
+2008_005045
+2008_005051
+2008_005055
+2008_005064
+2008_005066
+2008_005071
+2008_005074
+2008_005078
+2008_005080
+2008_005081
+2008_005082
+2008_005084
+2008_005088
+2008_005090
+2008_005094
+2008_005101
+2008_005108
+2008_005115
+2008_005127
+2008_005132
+2008_005133
+2008_005134
+2008_005136
+2008_005137
+2008_005146
+2008_005150
+2008_005158
+2008_005159
+2008_005168
+2008_005171
+2008_005172
+2008_005174
+2008_005178
+2008_005186
+2008_005193
+2008_005196
+2008_005201
+2008_005209
+2008_005213
+2008_005214
+2008_005216
+2008_005218
+2008_005220
+2008_005221
+2008_005231
+2008_005234
+2008_005236
+2008_005240
+2008_005247
+2008_005248
+2008_005250
+2008_005266
+2008_005269
+2008_005271
+2008_005279
+2008_005281
+2008_005283
+2008_005294
+2008_005295
+2008_005296
+2008_005297
+2008_005300
+2008_005303
+2008_005310
+2008_005315
+2008_005321
+2008_005324
+2008_005325
+2008_005329
+2008_005331
+2008_005333
+2008_005336
+2008_005342
+2008_005345
+2008_005349
+2008_005350
+2008_005354
+2008_005357
+2008_005362
+2008_005363
+2008_005365
+2008_005367
+2008_005375
+2008_005376
+2008_005380
+2008_005382
+2008_005386
+2008_005389
+2008_005395
+2008_005396
+2008_005400
+2008_005405
+2008_005408
+2008_005412
+2008_005414
+2008_005415
+2008_005429
+2008_005433
+2008_005443
+2008_005449
+2008_005451
+2008_005456
+2008_005463
+2008_005465
+2008_005473
+2008_005477
+2008_005484
+2008_005491
+2008_005494
+2008_005496
+2008_005500
+2008_005502
+2008_005505
+2008_005507
+2008_005512
+2008_005514
+2008_005517
+2008_005519
+2008_005521
+2008_005523
+2008_005526
+2008_005527
+2008_005531
+2008_005536
+2008_005541
+2008_005549
+2008_005558
+2008_005560
+2008_005561
+2008_005567
+2008_005569
+2008_005570
+2008_005572
+2008_005584
+2008_005589
+2008_005591
+2008_005593
+2008_005600
+2008_005603
+2008_005609
+2008_005610
+2008_005616
+2008_005618
+2008_005623
+2008_005625
+2008_005626
+2008_005634
+2008_005636
+2008_005639
+2008_005641
+2008_005650
+2008_005653
+2008_005656
+2008_005668
+2008_005673
+2008_005675
+2008_005678
+2008_005679
+2008_005682
+2008_005683
+2008_005698
+2008_005705
+2008_005706
+2008_005707
+2008_005713
+2008_005714
+2008_005716
+2008_005719
+2008_005724
+2008_005728
+2008_005736
+2008_005737
+2008_005739
+2008_005742
+2008_005747
+2008_005752
+2008_005757
+2008_005758
+2008_005761
+2008_005767
+2008_005770
+2008_005780
+2008_005791
+2008_005794
+2008_005800
+2008_005803
+2008_005810
+2008_005817
+2008_005818
+2008_005822
+2008_005823
+2008_005832
+2008_005834
+2008_005839
+2008_005843
+2008_005845
+2008_005850
+2008_005853
+2008_005855
+2008_005856
+2008_005867
+2008_005871
+2008_005873
+2008_005874
+2008_005878
+2008_005882
+2008_005890
+2008_005891
+2008_005893
+2008_005897
+2008_005902
+2008_005903
+2008_005905
+2008_005916
+2008_005921
+2008_005923
+2008_005926
+2008_005929
+2008_005935
+2008_005937
+2008_005938
+2008_005945
+2008_005954
+2008_005956
+2008_005959
+2008_005960
+2008_005967
+2008_005968
+2008_005972
+2008_005976
+2008_005979
+2008_005982
+2008_005991
+2008_005997
+2008_006000
+2008_006004
+2008_006014
+2008_006020
+2008_006032
+2008_006039
+2008_006046
+2008_006049
+2008_006062
+2008_006064
+2008_006065
+2008_006067
+2008_006070
+2008_006074
+2008_006076
+2008_006078
+2008_006081
+2008_006085
+2008_006090
+2008_006092
+2008_006096
+2008_006099
+2008_006102
+2008_006111
+2008_006119
+2008_006121
+2008_006124
+2008_006128
+2008_006129
+2008_006133
+2008_006135
+2008_006136
+2008_006140
+2008_006145
+2008_006152
+2008_006158
+2008_006164
+2008_006170
+2008_006181
+2008_006182
+2008_006186
+2008_006188
+2008_006192
+2008_006194
+2008_006195
+2008_006210
+2008_006211
+2008_006213
+2008_006215
+2008_006220
+2008_006221
+2008_006224
+2008_006225
+2008_006232
+2008_006234
+2008_006235
+2008_006242
+2008_006244
+2008_006249
+2008_006250
+2008_006253
+2008_006256
+2008_006257
+2008_006258
+2008_006265
+2008_006271
+2008_006272
+2008_006273
+2008_006276
+2008_006280
+2008_006281
+2008_006289
+2008_006294
+2008_006295
+2008_006300
+2008_006315
+2008_006317
+2008_006323
+2008_006329
+2008_006331
+2008_006335
+2008_006336
+2008_006339
+2008_006345
+2008_006349
+2008_006350
+2008_006351
+2008_006353
+2008_006355
+2008_006361
+2008_006364
+2008_006365
+2008_006369
+2008_006370
+2008_006373
+2008_006376
+2008_006384
+2008_006386
+2008_006387
+2008_006389
+2008_006390
+2008_006400
+2008_006401
+2008_006404
+2008_006410
+2008_006417
+2008_006419
+2008_006421
+2008_006427
+2008_006430
+2008_006432
+2008_006433
+2008_006434
+2008_006436
+2008_006447
+2008_006448
+2008_006461
+2008_006462
+2008_006470
+2008_006474
+2008_006475
+2008_006481
+2008_006482
+2008_006483
+2008_006488
+2008_006490
+2008_006491
+2008_006496
+2008_006497
+2008_006500
+2008_006506
+2008_006509
+2008_006511
+2008_006512
+2008_006522
+2008_006538
+2008_006540
+2008_006543
+2008_006546
+2008_006549
+2008_006558
+2008_006561
+2008_006562
+2008_006564
+2008_006566
+2008_006567
+2008_006570
+2008_006578
+2008_006579
+2008_006585
+2008_006586
+2008_006598
+2008_006599
+2008_006602
+2008_006606
+2008_006610
+2008_006613
+2008_006619
+2008_006623
+2008_006624
+2008_006625
+2008_006626
+2008_006629
+2008_006634
+2008_006637
+2008_006638
+2008_006641
+2008_006645
+2008_006649
+2008_006650
+2008_006654
+2008_006655
+2008_006657
+2008_006663
+2008_006667
+2008_006668
+2008_006677
+2008_006682
+2008_006691
+2008_006692
+2008_006700
+2008_006705
+2008_006712
+2008_006715
+2008_006717
+2008_006718
+2008_006719
+2008_006720
+2008_006724
+2008_006728
+2008_006730
+2008_006733
+2008_006737
+2008_006748
+2008_006750
+2008_006751
+2008_006753
+2008_006761
+2008_006762
+2008_006764
+2008_006767
+2008_006778
+2008_006785
+2008_006802
+2008_006807
+2008_006808
+2008_006810
+2008_006818
+2008_006819
+2008_006820
+2008_006827
+2008_006832
+2008_006834
+2008_006843
+2008_006847
+2008_006857
+2008_006864
+2008_006865
+2008_006868
+2008_006872
+2008_006873
+2008_006877
+2008_006879
+2008_006881
+2008_006882
+2008_006889
+2008_006898
+2008_006902
+2008_006903
+2008_006908
+2008_006909
+2008_006910
+2008_006919
+2008_006920
+2008_006921
+2008_006923
+2008_006926
+2008_006933
+2008_006936
+2008_006946
+2008_006950
+2008_006953
+2008_006954
+2008_006960
+2008_006961
+2008_006962
+2008_006965
+2008_006969
+2008_006973
+2008_006992
+2008_007003
+2008_007004
+2008_007009
+2008_007011
+2008_007012
+2008_007014
+2008_007022
+2008_007026
+2008_007028
+2008_007030
+2008_007038
+2008_007039
+2008_007043
+2008_007045
+2008_007054
+2008_007058
+2008_007060
+2008_007061
+2008_007069
+2008_007073
+2008_007075
+2008_007076
+2008_007081
+2008_007082
+2008_007085
+2008_007090
+2008_007095
+2008_007097
+2008_007098
+2008_007101
+2008_007106
+2008_007115
+2008_007118
+2008_007124
+2008_007129
+2008_007131
+2008_007138
+2008_007142
+2008_007145
+2008_007146
+2008_007147
+2008_007151
+2008_007156
+2008_007161
+2008_007165
+2008_007168
+2008_007169
+2008_007179
+2008_007185
+2008_007197
+2008_007201
+2008_007205
+2008_007208
+2008_007211
+2008_007217
+2008_007218
+2008_007221
+2008_007223
+2008_007226
+2008_007236
+2008_007237
+2008_007239
+2008_007242
+2008_007245
+2008_007246
+2008_007252
+2008_007254
+2008_007260
+2008_007261
+2008_007265
+2008_007274
+2008_007280
+2008_007281
+2008_007286
+2008_007289
+2008_007291
+2008_007298
+2008_007307
+2008_007312
+2008_007313
+2008_007320
+2008_007321
+2008_007325
+2008_007335
+2008_007343
+2008_007346
+2008_007353
+2008_007356
+2008_007357
+2008_007361
+2008_007363
+2008_007364
+2008_007375
+2008_007382
+2008_007383
+2008_007388
+2008_007394
+2008_007397
+2008_007410
+2008_007421
+2008_007423
+2008_007424
+2008_007425
+2008_007428
+2008_007432
+2008_007433
+2008_007438
+2008_007442
+2008_007443
+2008_007444
+2008_007448
+2008_007456
+2008_007465
+2008_007469
+2008_007470
+2008_007471
+2008_007472
+2008_007473
+2008_007477
+2008_007485
+2008_007486
+2008_007491
+2008_007496
+2008_007500
+2008_007504
+2008_007509
+2008_007510
+2008_007511
+2008_007515
+2008_007519
+2008_007524
+2008_007528
+2008_007533
+2008_007537
+2008_007544
+2008_007546
+2008_007556
+2008_007559
+2008_007565
+2008_007573
+2008_007576
+2008_007581
+2008_007584
+2008_007588
+2008_007589
+2008_007593
+2008_007597
+2008_007604
+2008_007608
+2008_007611
+2008_007613
+2008_007621
+2008_007625
+2008_007629
+2008_007630
+2008_007640
+2008_007641
+2008_007646
+2008_007648
+2008_007653
+2008_007660
+2008_007664
+2008_007665
+2008_007666
+2008_007675
+2008_007682
+2008_007683
+2008_007691
+2008_007692
+2008_007696
+2008_007697
+2008_007698
+2008_007701
+2008_007709
+2008_007710
+2008_007717
+2008_007724
+2008_007726
+2008_007730
+2008_007742
+2008_007746
+2008_007748
+2008_007750
+2008_007752
+2008_007755
+2008_007758
+2008_007759
+2008_007761
+2008_007770
+2008_007777
+2008_007779
+2008_007780
+2008_007781
+2008_007786
+2008_007788
+2008_007789
+2008_007805
+2008_007812
+2008_007817
+2008_007825
+2008_007829
+2008_007833
+2008_007835
+2008_007837
+2008_007840
+2008_007842
+2008_007843
+2008_007848
+2008_007852
+2008_007858
+2008_007861
+2008_007864
+2008_007869
+2008_007870
+2008_007873
+2008_007877
+2008_007879
+2008_007882
+2008_007883
+2008_007891
+2008_007895
+2008_007897
+2008_007904
+2008_007907
+2008_007909
+2008_007912
+2008_007913
+2008_007916
+2008_007918
+2008_007928
+2008_007937
+2008_007938
+2008_007940
+2008_007941
+2008_007947
+2008_007950
+2008_007953
+2008_007962
+2008_007969
+2008_007973
+2008_007975
+2008_007977
+2008_007985
+2008_007987
+2008_007988
+2008_007990
+2008_007997
+2008_007998
+2008_007999
+2008_008002
+2008_008004
+2008_008007
+2008_008012
+2008_008018
+2008_008020
+2008_008021
+2008_008028
+2008_008031
+2008_008034
+2008_008037
+2008_008043
+2008_008048
+2008_008058
+2008_008064
+2008_008070
+2008_008072
+2008_008073
+2008_008074
+2008_008080
+2008_008083
+2008_008092
+2008_008095
+2008_008097
+2008_008098
+2008_008106
+2008_008112
+2008_008116
+2008_008121
+2008_008122
+2008_008125
+2008_008132
+2008_008147
+2008_008148
+2008_008150
+2008_008152
+2008_008154
+2008_008162
+2008_008166
+2008_008169
+2008_008170
+2008_008176
+2008_008180
+2008_008184
+2008_008193
+2008_008194
+2008_008197
+2008_008199
+2008_008200
+2008_008206
+2008_008211
+2008_008212
+2008_008215
+2008_008218
+2008_008220
+2008_008223
+2008_008227
+2008_008229
+2008_008237
+2008_008242
+2008_008247
+2008_008262
+2008_008263
+2008_008266
+2008_008274
+2008_008275
+2008_008276
+2008_008281
+2008_008287
+2008_008288
+2008_008294
+2008_008300
+2008_008309
+2008_008315
+2008_008319
+2008_008321
+2008_008323
+2008_008324
+2008_008325
+2008_008330
+2008_008338
+2008_008342
+2008_008343
+2008_008344
+2008_008345
+2008_008347
+2008_008356
+2008_008363
+2008_008364
+2008_008366
+2008_008368
+2008_008370
+2008_008382
+2008_008384
+2008_008391
+2008_008402
+2008_008403
+2008_008404
+2008_008410
+2008_008411
+2008_008416
+2008_008423
+2008_008428
+2008_008431
+2008_008432
+2008_008440
+2008_008447
+2008_008455
+2008_008462
+2008_008464
+2008_008471
+2008_008476
+2008_008479
+2008_008480
+2008_008482
+2008_008487
+2008_008490
+2008_008496
+2008_008497
+2008_008507
+2008_008508
+2008_008511
+2008_008517
+2008_008521
+2008_008522
+2008_008523
+2008_008525
+2008_008526
+2008_008528
+2008_008530
+2008_008533
+2008_008541
+2008_008544
+2008_008545
+2008_008546
+2008_008547
+2008_008549
+2008_008550
+2008_008560
+2008_008567
+2008_008572
+2008_008578
+2008_008579
+2008_008583
+2008_008589
+2008_008590
+2008_008591
+2008_008593
+2008_008600
+2008_008601
+2008_008607
+2008_008608
+2008_008616
+2008_008618
+2008_008623
+2008_008624
+2008_008635
+2008_008637
+2008_008641
+2008_008642
+2008_008649
+2008_008654
+2008_008665
+2008_008666
+2008_008668
+2008_008671
+2008_008673
+2008_008674
+2008_008681
+2008_008685
+2008_008689
+2008_008691
+2008_008694
+2008_008696
+2008_008697
+2008_008701
+2008_008706
+2008_008707
+2008_008714
+2008_008717
+2008_008719
+2008_008725
+2008_008735
+2008_008744
+2008_008745
+2008_008748
+2008_008749
+2008_008757
+2008_008770
+2008_008773
+2009_000006
+2009_000010
+2009_000014
+2009_000015
+2009_000016
+2009_000021
+2009_000027
+2009_000028
+2009_000029
+2009_000030
+2009_000040
+2009_000042
+2009_000052
+2009_000054
+2009_000056
+2009_000058
+2009_000059
+2009_000073
+2009_000082
+2009_000085
+2009_000088
+2009_000091
+2009_000100
+2009_000103
+2009_000104
+2009_000105
+2009_000109
+2009_000119
+2009_000120
+2009_000122
+2009_000128
+2009_000130
+2009_000131
+2009_000132
+2009_000133
+2009_000135
+2009_000137
+2009_000140
+2009_000141
+2009_000145
+2009_000150
+2009_000151
+2009_000159
+2009_000160
+2009_000161
+2009_000164
+2009_000168
+2009_000176
+2009_000177
+2009_000188
+2009_000195
+2009_000197
+2009_000203
+2009_000209
+2009_000217
+2009_000218
+2009_000223
+2009_000227
+2009_000229
+2009_000232
+2009_000233
+2009_000237
+2009_000239
+2009_000248
+2009_000250
+2009_000251
+2009_000253
+2009_000268
+2009_000277
+2009_000280
+2009_000281
+2009_000285
+2009_000287
+2009_000289
+2009_000290
+2009_000303
+2009_000317
+2009_000320
+2009_000322
+2009_000327
+2009_000336
+2009_000339
+2009_000340
+2009_000341
+2009_000343
+2009_000344
+2009_000347
+2009_000350
+2009_000367
+2009_000375
+2009_000377
+2009_000379
+2009_000385
+2009_000390
+2009_000393
+2009_000400
+2009_000405
+2009_000408
+2009_000409
+2009_000416
+2009_000419
+2009_000420
+2009_000438
+2009_000439
+2009_000443
+2009_000444
+2009_000445
+2009_000449
+2009_000452
+2009_000454
+2009_000463
+2009_000464
+2009_000471
+2009_000474
+2009_000476
+2009_000477
+2009_000486
+2009_000491
+2009_000493
+2009_000494
+2009_000500
+2009_000502
+2009_000503
+2009_000504
+2009_000505
+2009_000515
+2009_000522
+2009_000525
+2009_000527
+2009_000529
+2009_000532
+2009_000535
+2009_000539
+2009_000544
+2009_000546
+2009_000547
+2009_000553
+2009_000557
+2009_000560
+2009_000562
+2009_000565
+2009_000567
+2009_000575
+2009_000576
+2009_000577
+2009_000579
+2009_000585
+2009_000586
+2009_000591
+2009_000592
+2009_000595
+2009_000599
+2009_000600
+2009_000602
+2009_000603
+2009_000604
+2009_000617
+2009_000626
+2009_000629
+2009_000632
+2009_000635
+2009_000636
+2009_000638
+2009_000642
+2009_000651
+2009_000653
+2009_000655
+2009_000662
+2009_000663
+2009_000672
+2009_000679
+2009_000684
+2009_000686
+2009_000690
+2009_000692
+2009_000694
+2009_000695
+2009_000696
+2009_000708
+2009_000709
+2009_000718
+2009_000720
+2009_000722
+2009_000737
+2009_000744
+2009_000745
+2009_000746
+2009_000748
+2009_000750
+2009_000752
+2009_000755
+2009_000757
+2009_000759
+2009_000768
+2009_000770
+2009_000774
+2009_000777
+2009_000789
+2009_000790
+2009_000793
+2009_000794
+2009_000796
+2009_000797
+2009_000801
+2009_000804
+2009_000805
+2009_000815
+2009_000816
+2009_000831
+2009_000833
+2009_000834
+2009_000848
+2009_000849
+2009_000854
+2009_000867
+2009_000869
+2009_000874
+2009_000882
+2009_000887
+2009_000889
+2009_000894
+2009_000895
+2009_000899
+2009_000902
+2009_000906
+2009_000910
+2009_000915
+2009_000920
+2009_000926
+2009_000927
+2009_000930
+2009_000932
+2009_000937
+2009_000938
+2009_000945
+2009_000953
+2009_000961
+2009_000962
+2009_000967
+2009_000969
+2009_000970
+2009_000973
+2009_000974
+2009_000975
+2009_000979
+2009_000980
+2009_000981
+2009_000987
+2009_000990
+2009_000996
+2009_001002
+2009_001009
+2009_001012
+2009_001013
+2009_001019
+2009_001027
+2009_001036
+2009_001037
+2009_001040
+2009_001042
+2009_001052
+2009_001056
+2009_001059
+2009_001068
+2009_001070
+2009_001074
+2009_001078
+2009_001079
+2009_001081
+2009_001085
+2009_001091
+2009_001095
+2009_001096
+2009_001098
+2009_001100
+2009_001102
+2009_001103
+2009_001104
+2009_001105
+2009_001107
+2009_001110
+2009_001111
+2009_001117
+2009_001124
+2009_001129
+2009_001133
+2009_001135
+2009_001137
+2009_001138
+2009_001140
+2009_001145
+2009_001146
+2009_001147
+2009_001151
+2009_001152
+2009_001153
+2009_001154
+2009_001159
+2009_001163
+2009_001172
+2009_001177
+2009_001180
+2009_001188
+2009_001190
+2009_001192
+2009_001197
+2009_001199
+2009_001201
+2009_001203
+2009_001205
+2009_001206
+2009_001208
+2009_001216
+2009_001217
+2009_001221
+2009_001224
+2009_001229
+2009_001230
+2009_001236
+2009_001237
+2009_001238
+2009_001241
+2009_001251
+2009_001253
+2009_001254
+2009_001260
+2009_001263
+2009_001264
+2009_001268
+2009_001270
+2009_001271
+2009_001282
+2009_001283
+2009_001285
+2009_001291
+2009_001301
+2009_001303
+2009_001305
+2009_001306
+2009_001308
+2009_001311
+2009_001312
+2009_001319
+2009_001323
+2009_001327
+2009_001328
+2009_001329
+2009_001339
+2009_001344
+2009_001354
+2009_001357
+2009_001359
+2009_001360
+2009_001364
+2009_001368
+2009_001369
+2009_001372
+2009_001374
+2009_001375
+2009_001376
+2009_001385
+2009_001388
+2009_001389
+2009_001390
+2009_001395
+2009_001403
+2009_001412
+2009_001422
+2009_001424
+2009_001434
+2009_001435
+2009_001443
+2009_001444
+2009_001446
+2009_001448
+2009_001450
+2009_001452
+2009_001453
+2009_001457
+2009_001462
+2009_001463
+2009_001466
+2009_001472
+2009_001474
+2009_001475
+2009_001476
+2009_001480
+2009_001481
+2009_001493
+2009_001494
+2009_001500
+2009_001502
+2009_001507
+2009_001508
+2009_001514
+2009_001516
+2009_001517
+2009_001537
+2009_001538
+2009_001541
+2009_001542
+2009_001544
+2009_001546
+2009_001550
+2009_001553
+2009_001555
+2009_001558
+2009_001566
+2009_001567
+2009_001570
+2009_001585
+2009_001589
+2009_001590
+2009_001595
+2009_001598
+2009_001602
+2009_001605
+2009_001608
+2009_001611
+2009_001612
+2009_001614
+2009_001615
+2009_001625
+2009_001636
+2009_001638
+2009_001640
+2009_001642
+2009_001651
+2009_001657
+2009_001660
+2009_001664
+2009_001670
+2009_001671
+2009_001674
+2009_001676
+2009_001677
+2009_001678
+2009_001689
+2009_001690
+2009_001693
+2009_001695
+2009_001704
+2009_001705
+2009_001706
+2009_001715
+2009_001719
+2009_001720
+2009_001724
+2009_001732
+2009_001734
+2009_001735
+2009_001740
+2009_001744
+2009_001746
+2009_001747
+2009_001749
+2009_001750
+2009_001751
+2009_001755
+2009_001770
+2009_001779
+2009_001781
+2009_001782
+2009_001783
+2009_001792
+2009_001798
+2009_001800
+2009_001801
+2009_001802
+2009_001806
+2009_001807
+2009_001809
+2009_001812
+2009_001817
+2009_001825
+2009_001826
+2009_001827
+2009_001828
+2009_001831
+2009_001837
+2009_001840
+2009_001846
+2009_001847
+2009_001856
+2009_001861
+2009_001865
+2009_001867
+2009_001868
+2009_001869
+2009_001871
+2009_001873
+2009_001874
+2009_001875
+2009_001884
+2009_001885
+2009_001888
+2009_001894
+2009_001897
+2009_001898
+2009_001902
+2009_001904
+2009_001908
+2009_001910
+2009_001917
+2009_001922
+2009_001926
+2009_001927
+2009_001933
+2009_001934
+2009_001937
+2009_001948
+2009_001952
+2009_001959
+2009_001960
+2009_001961
+2009_001962
+2009_001964
+2009_001972
+2009_001975
+2009_001990
+2009_001994
+2009_001997
+2009_001999
+2009_002000
+2009_002010
+2009_002018
+2009_002019
+2009_002037
+2009_002040
+2009_002044
+2009_002052
+2009_002054
+2009_002057
+2009_002060
+2009_002064
+2009_002066
+2009_002072
+2009_002077
+2009_002083
+2009_002086
+2009_002088
+2009_002089
+2009_002093
+2009_002096
+2009_002098
+2009_002099
+2009_002103
+2009_002104
+2009_002105
+2009_002107
+2009_002112
+2009_002116
+2009_002117
+2009_002118
+2009_002119
+2009_002120
+2009_002123
+2009_002126
+2009_002129
+2009_002133
+2009_002145
+2009_002146
+2009_002147
+2009_002149
+2009_002151
+2009_002152
+2009_002153
+2009_002173
+2009_002176
+2009_002180
+2009_002182
+2009_002192
+2009_002193
+2009_002197
+2009_002198
+2009_002203
+2009_002204
+2009_002214
+2009_002216
+2009_002225
+2009_002229
+2009_002235
+2009_002236
+2009_002240
+2009_002245
+2009_002253
+2009_002254
+2009_002256
+2009_002258
+2009_002259
+2009_002262
+2009_002264
+2009_002271
+2009_002273
+2009_002274
+2009_002281
+2009_002285
+2009_002289
+2009_002297
+2009_002298
+2009_002299
+2009_002301
+2009_002311
+2009_002312
+2009_002314
+2009_002324
+2009_002326
+2009_002331
+2009_002338
+2009_002339
+2009_002343
+2009_002348
+2009_002352
+2009_002358
+2009_002362
+2009_002371
+2009_002376
+2009_002377
+2009_002381
+2009_002386
+2009_002387
+2009_002388
+2009_002391
+2009_002397
+2009_002404
+2009_002406
+2009_002408
+2009_002409
+2009_002416
+2009_002419
+2009_002422
+2009_002423
+2009_002424
+2009_002425
+2009_002429
+2009_002431
+2009_002434
+2009_002438
+2009_002439
+2009_002443
+2009_002448
+2009_002452
+2009_002456
+2009_002460
+2009_002472
+2009_002504
+2009_002505
+2009_002506
+2009_002514
+2009_002519
+2009_002522
+2009_002523
+2009_002530
+2009_002536
+2009_002542
+2009_002543
+2009_002553
+2009_002556
+2009_002557
+2009_002558
+2009_002559
+2009_002561
+2009_002565
+2009_002567
+2009_002577
+2009_002579
+2009_002585
+2009_002586
+2009_002588
+2009_002595
+2009_002597
+2009_002599
+2009_002605
+2009_002611
+2009_002612
+2009_002613
+2009_002615
+2009_002616
+2009_002620
+2009_002621
+2009_002624
+2009_002625
+2009_002626
+2009_002628
+2009_002629
+2009_002648
+2009_002652
+2009_002659
+2009_002662
+2009_002671
+2009_002672
+2009_002674
+2009_002676
+2009_002685
+2009_002688
+2009_002689
+2009_002695
+2009_002697
+2009_002703
+2009_002704
+2009_002705
+2009_002710
+2009_002713
+2009_002714
+2009_002715
+2009_002719
+2009_002725
+2009_002728
+2009_002734
+2009_002746
+2009_002750
+2009_002758
+2009_002759
+2009_002763
+2009_002764
+2009_002770
+2009_002780
+2009_002784
+2009_002789
+2009_002791
+2009_002792
+2009_002798
+2009_002799
+2009_002813
+2009_002814
+2009_002817
+2009_002820
+2009_002824
+2009_002827
+2009_002831
+2009_002835
+2009_002842
+2009_002843
+2009_002844
+2009_002845
+2009_002847
+2009_002849
+2009_002850
+2009_002851
+2009_002853
+2009_002855
+2009_002862
+2009_002867
+2009_002869
+2009_002872
+2009_002879
+2009_002885
+2009_002890
+2009_002893
+2009_002897
+2009_002901
+2009_002908
+2009_002912
+2009_002914
+2009_002917
+2009_002921
+2009_002932
+2009_002933
+2009_002935
+2009_002937
+2009_002946
+2009_002947
+2009_002952
+2009_002954
+2009_002955
+2009_002957
+2009_002958
+2009_002961
+2009_002970
+2009_002971
+2009_002972
+2009_002976
+2009_002980
+2009_002983
+2009_002984
+2009_002988
+2009_002993
+2009_002999
+2009_003000
+2009_003002
+2009_003006
+2009_003007
+2009_003012
+2009_003019
+2009_003032
+2009_003034
+2009_003035
+2009_003039
+2009_003042
+2009_003053
+2009_003054
+2009_003056
+2009_003064
+2009_003066
+2009_003067
+2009_003068
+2009_003075
+2009_003077
+2009_003078
+2009_003082
+2009_003087
+2009_003088
+2009_003090
+2009_003091
+2009_003093
+2009_003095
+2009_003107
+2009_003108
+2009_003109
+2009_003115
+2009_003116
+2009_003118
+2009_003127
+2009_003138
+2009_003142
+2009_003146
+2009_003147
+2009_003155
+2009_003156
+2009_003157
+2009_003164
+2009_003165
+2009_003166
+2009_003168
+2009_003172
+2009_003173
+2009_003175
+2009_003187
+2009_003200
+2009_003208
+2009_003209
+2009_003218
+2009_003219
+2009_003222
+2009_003225
+2009_003229
+2009_003232
+2009_003233
+2009_003234
+2009_003249
+2009_003253
+2009_003257
+2009_003261
+2009_003265
+2009_003267
+2009_003272
+2009_003277
+2009_003285
+2009_003290
+2009_003309
+2009_003310
+2009_003315
+2009_003316
+2009_003317
+2009_003326
+2009_003327
+2009_003333
+2009_003338
+2009_003340
+2009_003345
+2009_003349
+2009_003350
+2009_003352
+2009_003353
+2009_003360
+2009_003361
+2009_003363
+2009_003365
+2009_003367
+2009_003369
+2009_003377
+2009_003381
+2009_003383
+2009_003384
+2009_003385
+2009_003386
+2009_003395
+2009_003396
+2009_003402
+2009_003407
+2009_003416
+2009_003419
+2009_003425
+2009_003430
+2009_003436
+2009_003443
+2009_003446
+2009_003447
+2009_003454
+2009_003455
+2009_003458
+2009_003459
+2009_003461
+2009_003468
+2009_003482
+2009_003488
+2009_003489
+2009_003490
+2009_003492
+2009_003497
+2009_003510
+2009_003511
+2009_003513
+2009_003519
+2009_003520
+2009_003522
+2009_003524
+2009_003531
+2009_003533
+2009_003534
+2009_003539
+2009_003540
+2009_003541
+2009_003545
+2009_003546
+2009_003555
+2009_003562
+2009_003563
+2009_003572
+2009_003577
+2009_003583
+2009_003594
+2009_003600
+2009_003601
+2009_003605
+2009_003608
+2009_003609
+2009_003613
+2009_003614
+2009_003624
+2009_003629
+2009_003634
+2009_003636
+2009_003639
+2009_003644
+2009_003646
+2009_003647
+2009_003652
+2009_003654
+2009_003657
+2009_003660
+2009_003663
+2009_003667
+2009_003668
+2009_003677
+2009_003683
+2009_003685
+2009_003688
+2009_003690
+2009_003694
+2009_003695
+2009_003697
+2009_003702
+2009_003705
+2009_003708
+2009_003709
+2009_003711
+2009_003717
+2009_003720
+2009_003722
+2009_003732
+2009_003734
+2009_003735
+2009_003736
+2009_003739
+2009_003743
+2009_003752
+2009_003753
+2009_003757
+2009_003760
+2009_003765
+2009_003768
+2009_003775
+2009_003783
+2009_003784
+2009_003786
+2009_003790
+2009_003793
+2009_003799
+2009_003801
+2009_003808
+2009_003815
+2009_003816
+2009_003818
+2009_003819
+2009_003820
+2009_003825
+2009_003827
+2009_003829
+2009_003837
+2009_003838
+2009_003843
+2009_003846
+2009_003848
+2009_003852
+2009_003860
+2009_003865
+2009_003867
+2009_003873
+2009_003883
+2009_003888
+2009_003896
+2009_003897
+2009_003900
+2009_003912
+2009_003913
+2009_003920
+2009_003921
+2009_003922
+2009_003933
+2009_003942
+2009_003956
+2009_003958
+2009_003961
+2009_003966
+2009_003974
+2009_003975
+2009_003976
+2009_003985
+2009_003993
+2009_003994
+2009_004002
+2009_004005
+2009_004007
+2009_004012
+2009_004018
+2009_004020
+2009_004023
+2009_004025
+2009_004037
+2009_004042
+2009_004055
+2009_004058
+2009_004069
+2009_004073
+2009_004074
+2009_004078
+2009_004082
+2009_004083
+2009_004088
+2009_004091
+2009_004094
+2009_004095
+2009_004096
+2009_004100
+2009_004103
+2009_004105
+2009_004109
+2009_004112
+2009_004117
+2009_004118
+2009_004121
+2009_004122
+2009_004133
+2009_004134
+2009_004139
+2009_004153
+2009_004154
+2009_004159
+2009_004162
+2009_004165
+2009_004168
+2009_004169
+2009_004171
+2009_004173
+2009_004174
+2009_004176
+2009_004177
+2009_004178
+2009_004179
+2009_004180
+2009_004181
+2009_004183
+2009_004186
+2009_004187
+2009_004191
+2009_004199
+2009_004200
+2009_004201
+2009_004202
+2009_004211
+2009_004212
+2009_004213
+2009_004218
+2009_004222
+2009_004225
+2009_004227
+2009_004228
+2009_004229
+2009_004231
+2009_004234
+2009_004244
+2009_004249
+2009_004261
+2009_004264
+2009_004271
+2009_004276
+2009_004278
+2009_004279
+2009_004283
+2009_004285
+2009_004289
+2009_004290
+2009_004295
+2009_004301
+2009_004308
+2009_004312
+2009_004315
+2009_004316
+2009_004317
+2009_004319
+2009_004322
+2009_004323
+2009_004327
+2009_004328
+2009_004334
+2009_004336
+2009_004338
+2009_004340
+2009_004341
+2009_004347
+2009_004351
+2009_004357
+2009_004358
+2009_004368
+2009_004369
+2009_004370
+2009_004371
+2009_004374
+2009_004375
+2009_004383
+2009_004392
+2009_004394
+2009_004397
+2009_004399
+2009_004406
+2009_004409
+2009_004417
+2009_004424
+2009_004425
+2009_004426
+2009_004429
+2009_004432
+2009_004434
+2009_004438
+2009_004442
+2009_004444
+2009_004445
+2009_004446
+2009_004449
+2009_004451
+2009_004452
+2009_004454
+2009_004464
+2009_004465
+2009_004475
+2009_004477
+2009_004479
+2009_004486
+2009_004492
+2009_004501
+2009_004503
+2009_004508
+2009_004511
+2009_004513
+2009_004514
+2009_004519
+2009_004527
+2009_004539
+2009_004545
+2009_004547
+2009_004554
+2009_004557
+2009_004560
+2009_004561
+2009_004562
+2009_004565
+2009_004570
+2009_004571
+2009_004572
+2009_004582
+2009_004593
+2009_004598
+2009_004606
+2009_004616
+2009_004619
+2009_004620
+2009_004626
+2009_004628
+2009_004631
+2009_004639
+2009_004642
+2009_004643
+2009_004647
+2009_004651
+2009_004652
+2009_004656
+2009_004661
+2009_004662
+2009_004667
+2009_004671
+2009_004674
+2009_004681
+2009_004683
+2009_004684
+2009_004688
+2009_004694
+2009_004701
+2009_004705
+2009_004708
+2009_004709
+2009_004710
+2009_004719
+2009_004723
+2009_004728
+2009_004731
+2009_004734
+2009_004737
+2009_004745
+2009_004756
+2009_004759
+2009_004760
+2009_004761
+2009_004764
+2009_004766
+2009_004771
+2009_004772
+2009_004779
+2009_004786
+2009_004787
+2009_004790
+2009_004794
+2009_004797
+2009_004798
+2009_004804
+2009_004805
+2009_004806
+2009_004813
+2009_004815
+2009_004817
+2009_004824
+2009_004829
+2009_004830
+2009_004831
+2009_004836
+2009_004839
+2009_004846
+2009_004847
+2009_004855
+2009_004871
+2009_004874
+2009_004877
+2009_004880
+2009_004887
+2009_004888
+2009_004890
+2009_004898
+2009_004901
+2009_004903
+2009_004904
+2009_004905
+2009_004907
+2009_004914
+2009_004919
+2009_004921
+2009_004926
+2009_004939
+2009_004943
+2009_004944
+2009_004945
+2009_004953
+2009_004958
+2009_004959
+2009_004962
+2009_004965
+2009_004972
+2009_004975
+2009_004977
+2009_004979
+2009_004980
+2009_004983
+2009_004984
+2009_004986
+2009_004990
+2009_004999
+2009_005000
+2009_005006
+2009_005015
+2009_005016
+2009_005024
+2009_005030
+2009_005031
+2009_005035
+2009_005037
+2009_005040
+2009_005042
+2009_005044
+2009_005045
+2009_005051
+2009_005055
+2009_005056
+2009_005057
+2009_005069
+2009_005070
+2009_005075
+2009_005076
+2009_005081
+2009_005084
+2009_005085
+2009_005094
+2009_005095
+2009_005102
+2009_005107
+2009_005118
+2009_005120
+2009_005126
+2009_005127
+2009_005128
+2009_005129
+2009_005130
+2009_005131
+2009_005133
+2009_005141
+2009_005142
+2009_005144
+2009_005145
+2009_005147
+2009_005154
+2009_005155
+2009_005160
+2009_005162
+2009_005163
+2009_005168
+2009_005170
+2009_005177
+2009_005181
+2009_005183
+2009_005191
+2009_005194
+2009_005198
+2009_005201
+2009_005218
+2009_005234
+2009_005236
+2009_005240
+2009_005246
+2009_005247
+2009_005251
+2009_005256
+2009_005263
+2009_005265
+2009_005269
+2009_005272
+2009_005278
+2009_005282
+2009_005287
+2009_005293
+2009_005297
+2009_005303
+2009_005307
+2009_005308
+2009_005311
+2010_000002
+2010_000009
+2010_000014
+2010_000018
+2010_000023
+2010_000026
+2010_000031
+2010_000043
+2010_000045
+2010_000048
+2010_000052
+2010_000055
+2010_000056
+2010_000061
+2010_000063
+2010_000067
+2010_000071
+2010_000073
+2010_000075
+2010_000076
+2010_000079
+2010_000080
+2010_000082
+2010_000089
+2010_000091
+2010_000103
+2010_000109
+2010_000111
+2010_000114
+2010_000117
+2010_000120
+2010_000124
+2010_000131
+2010_000132
+2010_000133
+2010_000136
+2010_000137
+2010_000138
+2010_000141
+2010_000148
+2010_000152
+2010_000157
+2010_000165
+2010_000169
+2010_000177
+2010_000182
+2010_000183
+2010_000187
+2010_000189
+2010_000190
+2010_000195
+2010_000198
+2010_000203
+2010_000204
+2010_000209
+2010_000222
+2010_000224
+2010_000227
+2010_000229
+2010_000233
+2010_000234
+2010_000244
+2010_000245
+2010_000248
+2010_000249
+2010_000250
+2010_000255
+2010_000263
+2010_000264
+2010_000269
+2010_000270
+2010_000276
+2010_000285
+2010_000293
+2010_000296
+2010_000299
+2010_000302
+2010_000303
+2010_000307
+2010_000310
+2010_000320
+2010_000323
+2010_000329
+2010_000337
+2010_000347
+2010_000356
+2010_000361
+2010_000362
+2010_000371
+2010_000377
+2010_000386
+2010_000388
+2010_000389
+2010_000392
+2010_000393
+2010_000394
+2010_000395
+2010_000404
+2010_000413
+2010_000415
+2010_000419
+2010_000420
+2010_000432
+2010_000436
+2010_000437
+2010_000439
+2010_000447
+2010_000448
+2010_000453
+2010_000458
+2010_000459
+2010_000463
+2010_000465
+2010_000466
+2010_000469
+2010_000473
+2010_000477
+2010_000480
+2010_000484
+2010_000488
+2010_000490
+2010_000492
+2010_000495
+2010_000498
+2010_000500
+2010_000503
+2010_000508
+2010_000511
+2010_000513
+2010_000519
+2010_000522
+2010_000527
+2010_000534
+2010_000538
+2010_000545
+2010_000549
+2010_000556
+2010_000557
+2010_000561
+2010_000564
+2010_000567
+2010_000568
+2010_000571
+2010_000576
+2010_000577
+2010_000578
+2010_000581
+2010_000588
+2010_000591
+2010_000601
+2010_000613
+2010_000616
+2010_000626
+2010_000630
+2010_000632
+2010_000641
+2010_000644
+2010_000645
+2010_000648
+2010_000651
+2010_000658
+2010_000661
+2010_000664
+2010_000667
+2010_000671
+2010_000674
+2010_000675
+2010_000678
+2010_000681
+2010_000685
+2010_000687
+2010_000688
+2010_000691
+2010_000694
+2010_000702
+2010_000707
+2010_000710
+2010_000715
+2010_000716
+2010_000717
+2010_000721
+2010_000723
+2010_000739
+2010_000740
+2010_000746
+2010_000747
+2010_000748
+2010_000750
+2010_000760
+2010_000765
+2010_000769
+2010_000770
+2010_000772
+2010_000773
+2010_000782
+2010_000785
+2010_000787
+2010_000799
+2010_000800
+2010_000803
+2010_000806
+2010_000807
+2010_000808
+2010_000810
+2010_000815
+2010_000827
+2010_000837
+2010_000838
+2010_000842
+2010_000847
+2010_000849
+2010_000855
+2010_000857
+2010_000860
+2010_000862
+2010_000863
+2010_000871
+2010_000872
+2010_000879
+2010_000885
+2010_000887
+2010_000891
+2010_000899
+2010_000908
+2010_000910
+2010_000912
+2010_000914
+2010_000920
+2010_000922
+2010_000926
+2010_000938
+2010_000939
+2010_000942
+2010_000954
+2010_000970
+2010_000971
+2010_000974
+2010_000978
+2010_000979
+2010_000983
+2010_000984
+2010_000986
+2010_000991
+2010_000994
+2010_000995
+2010_001002
+2010_001012
+2010_001013
+2010_001020
+2010_001023
+2010_001025
+2010_001039
+2010_001043
+2010_001044
+2010_001054
+2010_001063
+2010_001066
+2010_001074
+2010_001076
+2010_001087
+2010_001092
+2010_001094
+2010_001098
+2010_001100
+2010_001103
+2010_001105
+2010_001106
+2010_001110
+2010_001111
+2010_001112
+2010_001113
+2010_001118
+2010_001120
+2010_001121
+2010_001123
+2010_001126
+2010_001131
+2010_001134
+2010_001139
+2010_001140
+2010_001142
+2010_001143
+2010_001148
+2010_001152
+2010_001154
+2010_001159
+2010_001160
+2010_001175
+2010_001177
+2010_001179
+2010_001183
+2010_001184
+2010_001185
+2010_001193
+2010_001195
+2010_001199
+2010_001205
+2010_001210
+2010_001211
+2010_001212
+2010_001224
+2010_001225
+2010_001237
+2010_001240
+2010_001245
+2010_001247
+2010_001250
+2010_001253
+2010_001254
+2010_001261
+2010_001271
+2010_001273
+2010_001274
+2010_001275
+2010_001277
+2010_001279
+2010_001282
+2010_001288
+2010_001289
+2010_001299
+2010_001310
+2010_001311
+2010_001312
+2010_001317
+2010_001320
+2010_001328
+2010_001329
+2010_001337
+2010_001338
+2010_001339
+2010_001344
+2010_001347
+2010_001356
+2010_001360
+2010_001361
+2010_001363
+2010_001366
+2010_001370
+2010_001372
+2010_001374
+2010_001383
+2010_001385
+2010_001386
+2010_001390
+2010_001395
+2010_001397
+2010_001399
+2010_001401
+2010_001402
+2010_001406
+2010_001408
+2010_001410
+2010_001413
+2010_001418
+2010_001422
+2010_001425
+2010_001430
+2010_001431
+2010_001433
+2010_001434
+2010_001435
+2010_001450
+2010_001456
+2010_001457
+2010_001458
+2010_001464
+2010_001465
+2010_001472
+2010_001478
+2010_001480
+2010_001481
+2010_001487
+2010_001489
+2010_001499
+2010_001503
+2010_001511
+2010_001514
+2010_001515
+2010_001529
+2010_001533
+2010_001537
+2010_001547
+2010_001550
+2010_001551
+2010_001552
+2010_001555
+2010_001560
+2010_001561
+2010_001562
+2010_001569
+2010_001572
+2010_001576
+2010_001580
+2010_001583
+2010_001590
+2010_001592
+2010_001594
+2010_001595
+2010_001596
+2010_001599
+2010_001602
+2010_001603
+2010_001607
+2010_001608
+2010_001618
+2010_001619
+2010_001626
+2010_001630
+2010_001638
+2010_001644
+2010_001647
+2010_001649
+2010_001650
+2010_001660
+2010_001665
+2010_001674
+2010_001676
+2010_001687
+2010_001689
+2010_001694
+2010_001698
+2010_001700
+2010_001706
+2010_001709
+2010_001710
+2010_001715
+2010_001718
+2010_001719
+2010_001726
+2010_001729
+2010_001732
+2010_001743
+2010_001744
+2010_001746
+2010_001747
+2010_001748
+2010_001753
+2010_001756
+2010_001759
+2010_001762
+2010_001770
+2010_001776
+2010_001780
+2010_001784
+2010_001785
+2010_001794
+2010_001795
+2010_001797
+2010_001801
+2010_001806
+2010_001807
+2010_001808
+2010_001810
+2010_001817
+2010_001841
+2010_001842
+2010_001846
+2010_001849
+2010_001850
+2010_001852
+2010_001853
+2010_001856
+2010_001858
+2010_001860
+2010_001864
+2010_001870
+2010_001881
+2010_001884
+2010_001885
+2010_001896
+2010_001899
+2010_001911
+2010_001919
+2010_001922
+2010_001923
+2010_001924
+2010_001931
+2010_001933
+2010_001934
+2010_001939
+2010_001940
+2010_001941
+2010_001944
+2010_001948
+2010_001957
+2010_001960
+2010_001970
+2010_001973
+2010_001974
+2010_001976
+2010_001978
+2010_001979
+2010_001980
+2010_001981
+2010_001982
+2010_001993
+2010_001994
+2010_002015
+2010_002018
+2010_002020
+2010_002023
+2010_002026
+2010_002032
+2010_002037
+2010_002039
+2010_002042
+2010_002044
+2010_002045
+2010_002047
+2010_002054
+2010_002055
+2010_002057
+2010_002065
+2010_002068
+2010_002070
+2010_002080
+2010_002095
+2010_002097
+2010_002104
+2010_002107
+2010_002118
+2010_002121
+2010_002127
+2010_002129
+2010_002130
+2010_002132
+2010_002136
+2010_002139
+2010_002141
+2010_002143
+2010_002149
+2010_002152
+2010_002154
+2010_002166
+2010_002168
+2010_002176
+2010_002177
+2010_002179
+2010_002180
+2010_002185
+2010_002191
+2010_002193
+2010_002203
+2010_002204
+2010_002207
+2010_002208
+2010_002215
+2010_002216
+2010_002218
+2010_002220
+2010_002221
+2010_002226
+2010_002227
+2010_002236
+2010_002242
+2010_002243
+2010_002248
+2010_002254
+2010_002263
+2010_002267
+2010_002274
+2010_002278
+2010_002286
+2010_002295
+2010_002299
+2010_002301
+2010_002309
+2010_002312
+2010_002318
+2010_002320
+2010_002327
+2010_002333
+2010_002338
+2010_002346
+2010_002349
+2010_002353
+2010_002356
+2010_002363
+2010_002364
+2010_002368
+2010_002369
+2010_002371
+2010_002374
+2010_002378
+2010_002379
+2010_002382
+2010_002387
+2010_002391
+2010_002392
+2010_002393
+2010_002399
+2010_002400
+2010_002410
+2010_002413
+2010_002418
+2010_002424
+2010_002425
+2010_002429
+2010_002431
+2010_002435
+2010_002438
+2010_002439
+2010_002440
+2010_002445
+2010_002452
+2010_002455
+2010_002456
+2010_002457
+2010_002459
+2010_002462
+2010_002469
+2010_002472
+2010_002475
+2010_002485
+2010_002487
+2010_002492
+2010_002496
+2010_002497
+2010_002498
+2010_002499
+2010_002501
+2010_002507
+2010_002509
+2010_002513
+2010_002520
+2010_002527
+2010_002529
+2010_002532
+2010_002537
+2010_002551
+2010_002552
+2010_002553
+2010_002556
+2010_002562
+2010_002567
+2010_002570
+2010_002573
+2010_002575
+2010_002577
+2010_002582
+2010_002583
+2010_002589
+2010_002592
+2010_002594
+2010_002614
+2010_002615
+2010_002616
+2010_002618
+2010_002620
+2010_002624
+2010_002625
+2010_002626
+2010_002628
+2010_002642
+2010_002644
+2010_002647
+2010_002653
+2010_002656
+2010_002659
+2010_002662
+2010_002665
+2010_002674
+2010_002675
+2010_002684
+2010_002686
+2010_002688
+2010_002692
+2010_002696
+2010_002697
+2010_002702
+2010_002708
+2010_002720
+2010_002722
+2010_002729
+2010_002733
+2010_002734
+2010_002742
+2010_002746
+2010_002747
+2010_002750
+2010_002752
+2010_002759
+2010_002760
+2010_002772
+2010_002778
+2010_002779
+2010_002781
+2010_002786
+2010_002794
+2010_002797
+2010_002801
+2010_002805
+2010_002811
+2010_002813
+2010_002815
+2010_002816
+2010_002820
+2010_002821
+2010_002830
+2010_002831
+2010_002834
+2010_002838
+2010_002839
+2010_002841
+2010_002842
+2010_002843
+2010_002844
+2010_002851
+2010_002855
+2010_002856
+2010_002857
+2010_002865
+2010_002870
+2010_002880
+2010_002884
+2010_002891
+2010_002892
+2010_002896
+2010_002899
+2010_002901
+2010_002903
+2010_002907
+2010_002909
+2010_002915
+2010_002917
+2010_002931
+2010_002935
+2010_002937
+2010_002938
+2010_002941
+2010_002946
+2010_002947
+2010_002948
+2010_002955
+2010_002962
+2010_002973
+2010_002976
+2010_002978
+2010_002979
+2010_002982
+2010_002987
+2010_002990
+2010_003002
+2010_003003
+2010_003007
+2010_003010
+2010_003011
+2010_003013
+2010_003017
+2010_003025
+2010_003027
+2010_003028
+2010_003032
+2010_003034
+2010_003035
+2010_003037
+2010_003044
+2010_003047
+2010_003050
+2010_003053
+2010_003055
+2010_003056
+2010_003057
+2010_003062
+2010_003077
+2010_003078
+2010_003084
+2010_003086
+2010_003088
+2010_003093
+2010_003094
+2010_003097
+2010_003101
+2010_003106
+2010_003108
+2010_003114
+2010_003115
+2010_003117
+2010_003119
+2010_003137
+2010_003138
+2010_003143
+2010_003148
+2010_003149
+2010_003151
+2010_003153
+2010_003157
+2010_003159
+2010_003162
+2010_003169
+2010_003170
+2010_003173
+2010_003174
+2010_003179
+2010_003185
+2010_003186
+2010_003191
+2010_003192
+2010_003197
+2010_003203
+2010_003204
+2010_003206
+2010_003218
+2010_003222
+2010_003227
+2010_003230
+2010_003238
+2010_003241
+2010_003250
+2010_003252
+2010_003255
+2010_003256
+2010_003259
+2010_003263
+2010_003264
+2010_003269
+2010_003274
+2010_003280
+2010_003283
+2010_003290
+2010_003291
+2010_003297
+2010_003300
+2010_003301
+2010_003304
+2010_003305
+2010_003309
+2010_003329
+2010_003332
+2010_003333
+2010_003337
+2010_003342
+2010_003343
+2010_003344
+2010_003345
+2010_003350
+2010_003351
+2010_003353
+2010_003355
+2010_003367
+2010_003370
+2010_003371
+2010_003372
+2010_003374
+2010_003380
+2010_003383
+2010_003384
+2010_003391
+2010_003395
+2010_003400
+2010_003405
+2010_003406
+2010_003415
+2010_003421
+2010_003432
+2010_003435
+2010_003436
+2010_003437
+2010_003439
+2010_003469
+2010_003474
+2010_003477
+2010_003478
+2010_003481
+2010_003483
+2010_003491
+2010_003507
+2010_003509
+2010_003512
+2010_003513
+2010_003526
+2010_003529
+2010_003534
+2010_003535
+2010_003538
+2010_003539
+2010_003546
+2010_003549
+2010_003551
+2010_003554
+2010_003556
+2010_003560
+2010_003567
+2010_003574
+2010_003576
+2010_003582
+2010_003592
+2010_003598
+2010_003599
+2010_003601
+2010_003604
+2010_003608
+2010_003612
+2010_003618
+2010_003625
+2010_003629
+2010_003634
+2010_003635
+2010_003643
+2010_003644
+2010_003648
+2010_003649
+2010_003651
+2010_003656
+2010_003665
+2010_003670
+2010_003671
+2010_003672
+2010_003674
+2010_003677
+2010_003680
+2010_003686
+2010_003689
+2010_003690
+2010_003696
+2010_003703
+2010_003714
+2010_003717
+2010_003719
+2010_003721
+2010_003725
+2010_003734
+2010_003736
+2010_003737
+2010_003743
+2010_003747
+2010_003752
+2010_003754
+2010_003770
+2010_003773
+2010_003784
+2010_003788
+2010_003789
+2010_003791
+2010_003798
+2010_003799
+2010_003804
+2010_003815
+2010_003816
+2010_003818
+2010_003821
+2010_003822
+2010_003825
+2010_003837
+2010_003844
+2010_003845
+2010_003856
+2010_003860
+2010_003864
+2010_003865
+2010_003871
+2010_003874
+2010_003875
+2010_003877
+2010_003884
+2010_003887
+2010_003891
+2010_003892
+2010_003893
+2010_003894
+2010_003897
+2010_003899
+2010_003900
+2010_003906
+2010_003910
+2010_003911
+2010_003914
+2010_003925
+2010_003929
+2010_003931
+2010_003937
+2010_003938
+2010_003945
+2010_003949
+2010_003950
+2010_003954
+2010_003957
+2010_003958
+2010_003974
+2010_003982
+2010_003987
+2010_003994
+2010_003995
+2010_003996
+2010_004002
+2010_004005
+2010_004007
+2010_004008
+2010_004009
+2010_004011
+2010_004014
+2010_004017
+2010_004025
+2010_004028
+2010_004029
+2010_004030
+2010_004033
+2010_004043
+2010_004045
+2010_004048
+2010_004052
+2010_004053
+2010_004059
+2010_004060
+2010_004061
+2010_004062
+2010_004065
+2010_004066
+2010_004069
+2010_004071
+2010_004072
+2010_004074
+2010_004075
+2010_004081
+2010_004084
+2010_004089
+2010_004092
+2010_004108
+2010_004109
+2010_004111
+2010_004116
+2010_004118
+2010_004119
+2010_004121
+2010_004123
+2010_004130
+2010_004133
+2010_004138
+2010_004144
+2010_004148
+2010_004154
+2010_004160
+2010_004162
+2010_004163
+2010_004168
+2010_004171
+2010_004172
+2010_004175
+2010_004180
+2010_004186
+2010_004191
+2010_004192
+2010_004197
+2010_004198
+2010_004204
+2010_004210
+2010_004216
+2010_004222
+2010_004223
+2010_004231
+2010_004239
+2010_004242
+2010_004244
+2010_004247
+2010_004248
+2010_004249
+2010_004252
+2010_004256
+2010_004258
+2010_004259
+2010_004264
+2010_004271
+2010_004275
+2010_004276
+2010_004282
+2010_004283
+2010_004288
+2010_004289
+2010_004295
+2010_004296
+2010_004301
+2010_004306
+2010_004307
+2010_004311
+2010_004325
+2010_004327
+2010_004332
+2010_004333
+2010_004336
+2010_004344
+2010_004346
+2010_004349
+2010_004357
+2010_004358
+2010_004360
+2010_004361
+2010_004363
+2010_004365
+2010_004366
+2010_004367
+2010_004368
+2010_004370
+2010_004371
+2010_004373
+2010_004385
+2010_004402
+2010_004412
+2010_004423
+2010_004429
+2010_004436
+2010_004441
+2010_004445
+2010_004448
+2010_004450
+2010_004451
+2010_004459
+2010_004466
+2010_004467
+2010_004476
+2010_004477
+2010_004478
+2010_004481
+2010_004491
+2010_004492
+2010_004493
+2010_004499
+2010_004501
+2010_004511
+2010_004514
+2010_004517
+2010_004518
+2010_004521
+2010_004523
+2010_004540
+2010_004546
+2010_004558
+2010_004560
+2010_004561
+2010_004569
+2010_004573
+2010_004575
+2010_004576
+2010_004577
+2010_004581
+2010_004591
+2010_004592
+2010_004594
+2010_004598
+2010_004600
+2010_004601
+2010_004604
+2010_004609
+2010_004616
+2010_004620
+2010_004621
+2010_004625
+2010_004631
+2010_004638
+2010_004646
+2010_004655
+2010_004656
+2010_004657
+2010_004660
+2010_004665
+2010_004666
+2010_004669
+2010_004676
+2010_004680
+2010_004683
+2010_004690
+2010_004694
+2010_004696
+2010_004698
+2010_004703
+2010_004704
+2010_004708
+2010_004710
+2010_004712
+2010_004717
+2010_004721
+2010_004726
+2010_004728
+2010_004729
+2010_004730
+2010_004738
+2010_004741
+2010_004749
+2010_004751
+2010_004760
+2010_004765
+2010_004766
+2010_004770
+2010_004773
+2010_004777
+2010_004782
+2010_004791
+2010_004793
+2010_004797
+2010_004805
+2010_004806
+2010_004807
+2010_004808
+2010_004812
+2010_004816
+2010_004822
+2010_004824
+2010_004826
+2010_004831
+2010_004832
+2010_004838
+2010_004841
+2010_004844
+2010_004847
+2010_004848
+2010_004852
+2010_004855
+2010_004871
+2010_004874
+2010_004878
+2010_004879
+2010_004888
+2010_004890
+2010_004896
+2010_004900
+2010_004910
+2010_004913
+2010_004916
+2010_004918
+2010_004922
+2010_004928
+2010_004933
+2010_004937
+2010_004938
+2010_004942
+2010_004943
+2010_004944
+2010_004945
+2010_004948
+2010_004950
+2010_004953
+2010_004959
+2010_004960
+2010_004962
+2010_004963
+2010_004966
+2010_004968
+2010_004970
+2010_004971
+2010_004973
+2010_004974
+2010_004983
+2010_004987
+2010_004991
+2010_004995
+2010_004997
+2010_005002
+2010_005011
+2010_005016
+2010_005017
+2010_005018
+2010_005019
+2010_005022
+2010_005028
+2010_005033
+2010_005041
+2010_005054
+2010_005055
+2010_005060
+2010_005062
+2010_005064
+2010_005068
+2010_005071
+2010_005072
+2010_005080
+2010_005090
+2010_005093
+2010_005094
+2010_005098
+2010_005099
+2010_005100
+2010_005101
+2010_005106
+2010_005110
+2010_005111
+2010_005119
+2010_005127
+2010_005128
+2010_005129
+2010_005133
+2010_005134
+2010_005147
+2010_005149
+2010_005155
+2010_005161
+2010_005170
+2010_005182
+2010_005183
+2010_005190
+2010_005193
+2010_005198
+2010_005199
+2010_005201
+2010_005202
+2010_005211
+2010_005213
+2010_005216
+2010_005217
+2010_005223
+2010_005229
+2010_005232
+2010_005236
+2010_005238
+2010_005241
+2010_005253
+2010_005257
+2010_005258
+2010_005260
+2010_005261
+2010_005266
+2010_005270
+2010_005273
+2010_005274
+2010_005275
+2010_005276
+2010_005277
+2010_005279
+2010_005297
+2010_005299
+2010_005301
+2010_005303
+2010_005306
+2010_005308
+2010_005309
+2010_005310
+2010_005312
+2010_005317
+2010_005318
+2010_005320
+2010_005349
+2010_005350
+2010_005352
+2010_005359
+2010_005361
+2010_005364
+2010_005365
+2010_005371
+2010_005376
+2010_005377
+2010_005384
+2010_005385
+2010_005386
+2010_005388
+2010_005389
+2010_005391
+2010_005393
+2010_005402
+2010_005403
+2010_005408
+2010_005409
+2010_005415
+2010_005417
+2010_005419
+2010_005426
+2010_005429
+2010_005434
+2010_005437
+2010_005442
+2010_005450
+2010_005457
+2010_005458
+2010_005462
+2010_005466
+2010_005468
+2010_005471
+2010_005475
+2010_005489
+2010_005492
+2010_005494
+2010_005497
+2010_005498
+2010_005500
+2010_005505
+2010_005506
+2010_005511
+2010_005512
+2010_005513
+2010_005518
+2010_005519
+2010_005522
+2010_005535
+2010_005536
+2010_005540
+2010_005546
+2010_005557
+2010_005559
+2010_005561
+2010_005565
+2010_005570
+2010_005571
+2010_005573
+2010_005578
+2010_005584
+2010_005585
+2010_005588
+2010_005591
+2010_005593
+2010_005595
+2010_005596
+2010_005597
+2010_005601
+2010_005603
+2010_005604
+2010_005608
+2010_005614
+2010_005615
+2010_005616
+2010_005619
+2010_005627
+2010_005628
+2010_005629
+2010_005640
+2010_005643
+2010_005646
+2010_005652
+2010_005663
+2010_005665
+2010_005668
+2010_005669
+2010_005670
+2010_005672
+2010_005678
+2010_005683
+2010_005684
+2010_005696
+2010_005700
+2010_005715
+2010_005716
+2010_005721
+2010_005723
+2010_005725
+2010_005732
+2010_005734
+2010_005735
+2010_005736
+2010_005740
+2010_005744
+2010_005746
+2010_005748
+2010_005750
+2010_005753
+2010_005755
+2010_005758
+2010_005770
+2010_005775
+2010_005776
+2010_005782
+2010_005785
+2010_005791
+2010_005794
+2010_005796
+2010_005800
+2010_005805
+2010_005807
+2010_005810
+2010_005816
+2010_005820
+2010_005821
+2010_005823
+2010_005825
+2010_005826
+2010_005830
+2010_005835
+2010_005836
+2010_005840
+2010_005841
+2010_005845
+2010_005847
+2010_005854
+2010_005855
+2010_005865
+2010_005867
+2010_005874
+2010_005875
+2010_005876
+2010_005891
+2010_005892
+2010_005898
+2010_005904
+2010_005906
+2010_005909
+2010_005919
+2010_005921
+2010_005927
+2010_005928
+2010_005929
+2010_005930
+2010_005932
+2010_005935
+2010_005942
+2010_005948
+2010_005949
+2010_005951
+2010_005952
+2010_005954
+2010_005957
+2010_005958
+2010_005959
+2010_005960
+2010_005967
+2010_005968
+2010_005972
+2010_005974
+2010_005975
+2010_005978
+2010_005982
+2010_005984
+2010_005985
+2010_005986
+2010_005987
+2010_005995
+2010_005996
+2010_006009
+2010_006012
+2010_006015
+2010_006023
+2010_006028
+2010_006040
+2010_006042
+2010_006050
+2010_006063
+2010_006066
+2010_006067
+2010_006073
+2010_006078
+2010_006079
+2011_000003
+2011_000006
+2011_000012
+2011_000017
+2011_000022
+2011_000025
+2011_000027
+2011_000028
+2011_000030
+2011_000041
+2011_000044
+2011_000048
+2011_000052
+2011_000053
+2011_000058
+2011_000068
+2011_000069
+2011_000072
+2011_000095
+2011_000105
+2011_000108
+2011_000116
+2011_000122
+2011_000137
+2011_000138
+2011_000145
+2011_000149
+2011_000152
+2011_000176
+2011_000181
+2011_000182
+2011_000192
+2011_000196
+2011_000197
+2011_000208
+2011_000216
+2011_000219
+2011_000220
+2011_000221
+2011_000222
+2011_000224
+2011_000228
+2011_000233
+2011_000241
+2011_000243
+2011_000249
+2011_000250
+2011_000252
+2011_000258
+2011_000267
+2011_000268
+2011_000269
+2011_000277
+2011_000278
+2011_000282
+2011_000285
+2011_000286
+2011_000290
+2011_000293
+2011_000297
+2011_000305
+2011_000317
+2011_000324
+2011_000329
+2011_000342
+2011_000343
+2011_000345
+2011_000347
+2011_000359
+2011_000361
+2011_000362
+2011_000370
+2011_000375
+2011_000376
+2011_000379
+2011_000382
+2011_000383
+2011_000385
+2011_000388
+2011_000392
+2011_000397
+2011_000398
+2011_000399
+2011_000400
+2011_000413
+2011_000416
+2011_000420
+2011_000428
+2011_000430
+2011_000434
+2011_000442
+2011_000444
+2011_000449
+2011_000450
+2011_000453
+2011_000454
+2011_000457
+2011_000461
+2011_000465
+2011_000468
+2011_000469
+2011_000472
+2011_000475
+2011_000485
+2011_000491
+2011_000492
+2011_000494
+2011_000496
+2011_000499
+2011_000502
+2011_000505
+2011_000509
+2011_000513
+2011_000520
+2011_000531
+2011_000534
+2011_000538
+2011_000542
+2011_000550
+2011_000551
+2011_000556
+2011_000558
+2011_000560
+2011_000565
+2011_000567
+2011_000572
+2011_000573
+2011_000577
+2011_000578
+2011_000579
+2011_000586
+2011_000589
+2011_000594
+2011_000596
+2011_000621
+2011_000628
+2011_000629
+2011_000631
+2011_000637
+2011_000641
+2011_000642
+2011_000646
+2011_000651
+2011_000652
+2011_000655
+2011_000657
+2011_000673
+2011_000675
+2011_000682
+2011_000684
+2011_000689
+2011_000692
+2011_000698
+2011_000701
+2011_000703
+2011_000704
+2011_000711
+2011_000713
+2011_000725
+2011_000730
+2011_000731
+2011_000748
+2011_000755
+2011_000757
+2011_000758
+2011_000759
+2011_000763
+2011_000768
+2011_000769
+2011_000771
+2011_000788
+2011_000790
+2011_000791
+2011_000793
+2011_000800
+2011_000804
+2011_000806
+2011_000815
+2011_000819
+2011_000820
+2011_000823
+2011_000827
+2011_000828
+2011_000829
+2011_000831
+2011_000834
+2011_000837
+2011_000839
+2011_000840
+2011_000845
+2011_000847
+2011_000848
+2011_000855
+2011_000858
+2011_000859
+2011_000875
+2011_000882
+2011_000885
+2011_000893
+2011_000895
+2011_000898
+2011_000899
+2011_000920
+2011_000922
+2011_000934
+2011_000940
+2011_000944
+2011_000947
+2011_000954
+2011_000971
+2011_000973
+2011_000975
+2011_000979
+2011_000981
+2011_000982
+2011_000983
+2011_000987
+2011_000991
+2011_000996
+2011_000997
+2011_000999
+2011_001001
+2011_001004
+2011_001009
+2011_001010
+2011_001011
+2011_001015
+2011_001016
+2011_001022
+2011_001023
+2011_001027
+2011_001028
+2011_001030
+2011_001031
+2011_001033
+2011_001034
+2011_001052
+2011_001055
+2011_001062
+2011_001066
+2011_001073
+2011_001079
+2011_001080
+2011_001091
+2011_001093
+2011_001097
+2011_001107
+2011_001117
+2011_001123
+2011_001127
+2011_001133
+2011_001134
+2011_001135
+2011_001136
+2011_001139
+2011_001144
+2011_001150
+2011_001153
+2011_001163
+2011_001166
+2011_001168
+2011_001169
+2011_001173
+2011_001175
+2011_001176
+2011_001188
+2011_001189
+2011_001192
+2011_001193
+2011_001198
+2011_001208
+2011_001211
+2011_001215
+2011_001216
+2011_001220
+2011_001227
+2011_001238
+2011_001240
+2011_001246
+2011_001253
+2011_001254
+2011_001255
+2011_001257
+2011_001259
+2011_001270
+2011_001272
+2011_001277
+2011_001285
+2011_001286
+2011_001302
+2011_001310
+2011_001318
+2011_001320
+2011_001323
+2011_001333
+2011_001336
+2011_001344
+2011_001354
+2011_001357
+2011_001369
+2011_001373
+2011_001381
+2011_001382
+2011_001384
+2011_001394
+2011_001400
+2011_001402
+2011_001411
+2011_001412
+2011_001414
+2011_001422
+2011_001424
+2011_001432
+2011_001449
+2011_001451
+2011_001455
+2011_001456
+2011_001463
+2011_001464
+2011_001466
+2011_001475
+2011_001476
+2011_001479
+2011_001480
+2011_001498
+2011_001503
+2011_001505
+2011_001510
+2011_001514
+2011_001519
+2011_001526
+2011_001532
+2011_001536
+2011_001537
+2011_001538
+2011_001542
+2011_001547
+2011_001549
+2011_001557
+2011_001560
+2011_001566
+2011_001571
+2011_001572
+2011_001582
+2011_001586
+2011_001599
+2011_001600
+2011_001602
+2011_001605
+2011_001606
+2011_001611
+2011_001616
+2011_001621
+2011_001622
+2011_001625
+2011_001629
+2011_001632
+2011_001643
+2011_001647
+2011_001649
+2011_001650
+2011_001652
+2011_001653
+2011_001656
+2011_001662
+2011_001663
+2011_001666
+2011_001671
+2011_001673
+2011_001679
+2011_001689
+2011_001694
+2011_001695
+2011_001698
+2011_001700
+2011_001710
+2011_001715
+2011_001716
+2011_001727
+2011_001730
+2011_001732
+2011_001733
+2011_001739
+2011_001740
+2011_001753
+2011_001754
+2011_001755
+2011_001764
+2011_001765
+2011_001766
+2011_001769
+2011_001776
+2011_001779
+2011_001789
+2011_001790
+2011_001791
+2011_001796
+2011_001799
+2011_001805
+2011_001810
+2011_001811
+2011_001826
+2011_001833
+2011_001840
+2011_001855
+2011_001866
+2011_001871
+2011_001872
+2011_001875
+2011_001884
+2011_001885
+2011_001886
+2011_001889
+2011_001891
+2011_001893
+2011_001895
+2011_001896
+2011_001901
+2011_001902
+2011_001904
+2011_001906
+2011_001920
+2011_001922
+2011_001924
+2011_001926
+2011_001928
+2011_001929
+2011_001930
+2011_001937
+2011_001938
+2011_001944
+2011_001949
+2011_001950
+2011_001952
+2011_001956
+2011_001959
+2011_001961
+2011_001964
+2011_001967
+2011_001971
+2011_001972
+2011_001974
+2011_001977
+2011_001987
+2011_001991
+2011_002005
+2011_002006
+2011_002012
+2011_002022
+2011_002027
+2011_002031
+2011_002034
+2011_002039
+2011_002046
+2011_002049
+2011_002050
+2011_002053
+2011_002055
+2011_002062
+2011_002063
+2011_002073
+2011_002085
+2011_002096
+2011_002097
+2011_002106
+2011_002107
+2011_002111
+2011_002113
+2011_002114
+2011_002119
+2011_002131
+2011_002134
+2011_002135
+2011_002142
+2011_002143
+2011_002144
+2011_002147
+2011_002148
+2011_002149
+2011_002167
+2011_002177
+2011_002179
+2011_002186
+2011_002189
+2011_002211
+2011_002218
+2011_002222
+2011_002224
+2011_002227
+2011_002228
+2011_002236
+2011_002237
+2011_002239
+2011_002245
+2011_002246
+2011_002251
+2011_002252
+2011_002253
+2011_002265
+2011_002268
+2011_002273
+2011_002278
+2011_002281
+2011_002284
+2011_002291
+2011_002300
+2011_002303
+2011_002318
+2011_002335
+2011_002341
+2011_002346
+2011_002347
+2011_002348
+2011_002350
+2011_002359
+2011_002381
+2011_002385
+2011_002387
+2011_002388
+2011_002389
+2011_002394
+2011_002397
+2011_002398
+2011_002402
+2011_002410
+2011_002413
+2011_002418
+2011_002419
+2011_002420
+2011_002421
+2011_002422
+2011_002433
+2011_002435
+2011_002436
+2011_002443
+2011_002447
+2011_002448
+2011_002455
+2011_002457
+2011_002458
+2011_002460
+2011_002461
+2011_002462
+2011_002464
+2011_002470
+2011_002474
+2011_002476
+2011_002484
+2011_002488
+2011_002492
+2011_002503
+2011_002504
+2011_002511
+2011_002514
+2011_002526
+2011_002528
+2011_002533
+2011_002543
+2011_002551
+2011_002552
+2011_002553
+2011_002554
+2011_002555
+2011_002559
+2011_002560
+2011_002561
+2011_002567
+2011_002568
+2011_002571
+2011_002584
+2011_002585
+2011_002590
+2011_002594
+2011_002598
+2011_002601
+2011_002606
+2011_002609
+2011_002614
+2011_002616
+2011_002618
+2011_002620
+2011_002636
+2011_002638
+2011_002649
+2011_002650
+2011_002652
+2011_002656
+2011_002657
+2011_002658
+2011_002661
+2011_002664
+2011_002673
+2011_002676
+2011_002677
+2011_002697
+2011_002706
+2011_002709
+2011_002715
+2011_002717
+2011_002719
+2011_002724
+2011_002726
+2011_002746
+2011_002748
+2011_002752
+2011_002756
+2011_002767
+2011_002770
+2011_002775
+2011_002776
+2011_002779
+2011_002780
+2011_002782
+2011_002790
+2011_002795
+2011_002798
+2011_002803
+2011_002808
+2011_002811
+2011_002814
+2011_002818
+2011_002821
+2011_002823
+2011_002826
+2011_002834
+2011_002842
+2011_002851
+2011_002852
+2011_002867
+2011_002872
+2011_002873
+2011_002881
+2011_002884
+2011_002889
+2011_002908
+2011_002911
+2011_002912
+2011_002913
+2011_002917
+2011_002920
+2011_002921
+2011_002924
+2011_002927
+2011_002930
+2011_002932
+2011_002935
+2011_002937
+2011_002938
+2011_002940
+2011_002942
+2011_002947
+2011_002949
+2011_002953
+2011_002956
+2011_002958
+2011_002965
+2011_002966
+2011_002969
+2011_002974
+2011_002979
+2011_002987
+2011_002988
+2011_003002
+2011_003005
+2011_003010
+2011_003016
+2011_003020
+2011_003025
+2011_003034
+2011_003038
+2011_003041
+2011_003044
+2011_003047
+2011_003048
+2011_003049
+2011_003054
+2011_003057
+2011_003063
+2011_003065
+2011_003066
+2011_003073
+2011_003074
+2011_003078
+2011_003081
+2011_003091
+2011_003109
+2011_003121
+2011_003124
+2011_003132
+2011_003134
+2011_003138
+2011_003141
+2011_003148
+2011_003150
+2011_003151
+2011_003154
+2011_003158
+2011_003159
+2011_003162
+2011_003171
+2011_003177
+2011_003183
+2011_003184
+2011_003187
+2011_003188
+2011_003192
+2011_003194
+2011_003216
+2011_003223
+2011_003230
+2011_003236
+2011_003238
+2011_003246
+2011_003247
+2011_003253
+2011_003255
+2011_003259
+2011_003274
+2011_003276
diff --git a/ImageSets/Main/trainval.txt b/ImageSets/Main/trainval.txt
new file mode 100644
index 0000000..2994c00
--- /dev/null
+++ b/ImageSets/Main/trainval.txt
@@ -0,0 +1,16551 @@
+000005
+000007
+000009
+000012
+000016
+000017
+000019
+000020
+000021
+000023
+000024
+000026
+000030
+000032
+000033
+000034
+000035
+000036
+000039
+000041
+000042
+000044
+000046
+000047
+000048
+000050
+000051
+000052
+000060
+000061
+000063
+000064
+000065
+000066
+000072
+000073
+000077
+000078
+000081
+000083
+000089
+000091
+000093
+000095
+000099
+000101
+000102
+000104
+000107
+000109
+000110
+000112
+000113
+000117
+000118
+000120
+000121
+000122
+000123
+000125
+000129
+000130
+000131
+000132
+000133
+000134
+000138
+000140
+000141
+000142
+000143
+000146
+000147
+000150
+000153
+000154
+000156
+000158
+000159
+000161
+000162
+000163
+000164
+000165
+000169
+000170
+000171
+000173
+000174
+000177
+000180
+000184
+000187
+000189
+000190
+000192
+000193
+000194
+000198
+000200
+000203
+000207
+000208
+000209
+000210
+000211
+000214
+000215
+000218
+000219
+000220
+000221
+000222
+000224
+000225
+000228
+000229
+000232
+000233
+000235
+000236
+000241
+000242
+000244
+000245
+000246
+000249
+000250
+000251
+000256
+000257
+000259
+000262
+000263
+000266
+000268
+000269
+000270
+000275
+000276
+000278
+000282
+000285
+000288
+000289
+000294
+000296
+000298
+000302
+000303
+000304
+000305
+000306
+000307
+000308
+000311
+000312
+000317
+000318
+000320
+000321
+000322
+000323
+000325
+000328
+000329
+000331
+000332
+000334
+000336
+000337
+000338
+000340
+000343
+000344
+000347
+000349
+000352
+000354
+000355
+000359
+000363
+000367
+000370
+000372
+000373
+000374
+000379
+000380
+000381
+000382
+000387
+000391
+000394
+000395
+000396
+000400
+000403
+000404
+000406
+000407
+000408
+000411
+000416
+000417
+000419
+000420
+000424
+000427
+000428
+000430
+000431
+000433
+000435
+000438
+000439
+000443
+000446
+000448
+000450
+000454
+000459
+000460
+000461
+000462
+000463
+000464
+000468
+000469
+000470
+000474
+000476
+000477
+000480
+000482
+000483
+000484
+000486
+000489
+000491
+000492
+000494
+000496
+000498
+000499
+000500
+000501
+000503
+000508
+000509
+000513
+000514
+000515
+000516
+000518
+000519
+000520
+000522
+000523
+000524
+000525
+000526
+000528
+000530
+000531
+000535
+000537
+000540
+000541
+000543
+000544
+000545
+000549
+000550
+000552
+000554
+000555
+000559
+000563
+000564
+000565
+000577
+000579
+000581
+000582
+000583
+000588
+000589
+000590
+000591
+000592
+000597
+000598
+000599
+000601
+000605
+000608
+000609
+000610
+000612
+000613
+000619
+000620
+000622
+000625
+000626
+000628
+000632
+000633
+000635
+000637
+000645
+000647
+000648
+000653
+000654
+000656
+000657
+000660
+000661
+000663
+000667
+000671
+000672
+000675
+000676
+000677
+000680
+000682
+000684
+000685
+000686
+000688
+000689
+000690
+000694
+000695
+000699
+000700
+000702
+000705
+000707
+000709
+000710
+000711
+000712
+000713
+000714
+000717
+000720
+000726
+000728
+000729
+000730
+000731
+000733
+000738
+000739
+000740
+000742
+000746
+000748
+000750
+000752
+000753
+000754
+000755
+000756
+000760
+000761
+000763
+000764
+000767
+000768
+000770
+000771
+000772
+000774
+000776
+000777
+000780
+000782
+000786
+000787
+000791
+000793
+000794
+000796
+000797
+000799
+000800
+000802
+000804
+000805
+000806
+000808
+000810
+000812
+000814
+000815
+000816
+000818
+000820
+000822
+000823
+000826
+000827
+000828
+000829
+000830
+000831
+000832
+000834
+000842
+000843
+000845
+000847
+000848
+000849
+000850
+000851
+000854
+000855
+000857
+000859
+000860
+000862
+000863
+000865
+000867
+000868
+000871
+000872
+000874
+000876
+000878
+000879
+000880
+000882
+000885
+000887
+000888
+000889
+000892
+000895
+000896
+000898
+000899
+000900
+000902
+000903
+000904
+000906
+000908
+000911
+000912
+000915
+000917
+000918
+000919
+000920
+000921
+000923
+000926
+000929
+000931
+000934
+000935
+000936
+000937
+000943
+000946
+000947
+000948
+000949
+000950
+000951
+000954
+000958
+000962
+000964
+000965
+000966
+000967
+000971
+000972
+000973
+000977
+000980
+000982
+000987
+000989
+000991
+000993
+000996
+000997
+000999
+001001
+001002
+001004
+001008
+001009
+001010
+001011
+001012
+001014
+001015
+001017
+001018
+001024
+001027
+001028
+001036
+001041
+001042
+001043
+001045
+001050
+001052
+001053
+001056
+001057
+001060
+001061
+001062
+001064
+001066
+001068
+001069
+001071
+001072
+001073
+001074
+001077
+001078
+001079
+001082
+001083
+001084
+001091
+001092
+001093
+001097
+001101
+001102
+001104
+001106
+001107
+001109
+001110
+001112
+001113
+001119
+001121
+001124
+001125
+001127
+001129
+001130
+001136
+001137
+001140
+001142
+001143
+001144
+001145
+001147
+001148
+001149
+001151
+001152
+001154
+001156
+001158
+001160
+001161
+001164
+001166
+001168
+001170
+001171
+001172
+001174
+001175
+001176
+001182
+001184
+001185
+001186
+001187
+001191
+001192
+001194
+001199
+001200
+001201
+001203
+001204
+001205
+001206
+001207
+001209
+001211
+001212
+001214
+001215
+001221
+001224
+001225
+001226
+001229
+001230
+001231
+001233
+001234
+001236
+001237
+001239
+001240
+001241
+001247
+001248
+001250
+001254
+001258
+001259
+001260
+001263
+001265
+001266
+001268
+001269
+001270
+001272
+001273
+001274
+001277
+001279
+001281
+001284
+001286
+001287
+001288
+001289
+001290
+001292
+001293
+001294
+001298
+001299
+001304
+001309
+001310
+001311
+001312
+001314
+001315
+001316
+001323
+001324
+001325
+001326
+001327
+001330
+001332
+001333
+001334
+001337
+001341
+001343
+001345
+001346
+001348
+001350
+001352
+001360
+001361
+001362
+001364
+001365
+001371
+001375
+001378
+001383
+001384
+001385
+001386
+001387
+001388
+001390
+001393
+001395
+001397
+001400
+001402
+001404
+001405
+001406
+001408
+001409
+001413
+001414
+001418
+001420
+001421
+001426
+001427
+001430
+001432
+001434
+001436
+001439
+001441
+001442
+001443
+001444
+001445
+001450
+001451
+001453
+001455
+001457
+001460
+001463
+001464
+001465
+001466
+001467
+001468
+001470
+001472
+001475
+001479
+001480
+001481
+001483
+001484
+001485
+001486
+001488
+001490
+001492
+001493
+001494
+001497
+001498
+001499
+001501
+001504
+001509
+001510
+001512
+001514
+001515
+001517
+001521
+001522
+001523
+001524
+001526
+001528
+001529
+001531
+001532
+001536
+001537
+001539
+001541
+001543
+001544
+001545
+001548
+001553
+001554
+001555
+001556
+001557
+001559
+001561
+001563
+001565
+001571
+001576
+001577
+001579
+001580
+001582
+001586
+001588
+001590
+001593
+001594
+001595
+001597
+001598
+001603
+001604
+001607
+001608
+001610
+001611
+001612
+001614
+001617
+001618
+001622
+001627
+001628
+001630
+001632
+001633
+001636
+001638
+001640
+001642
+001643
+001647
+001649
+001650
+001651
+001653
+001654
+001661
+001662
+001669
+001673
+001675
+001676
+001677
+001678
+001680
+001682
+001683
+001684
+001685
+001686
+001688
+001689
+001690
+001691
+001693
+001699
+001707
+001708
+001711
+001713
+001714
+001717
+001718
+001721
+001723
+001724
+001725
+001726
+001727
+001729
+001730
+001732
+001733
+001734
+001738
+001739
+001741
+001746
+001747
+001749
+001750
+001752
+001754
+001755
+001756
+001758
+001759
+001761
+001765
+001766
+001768
+001771
+001772
+001775
+001777
+001778
+001780
+001782
+001784
+001785
+001787
+001789
+001793
+001795
+001797
+001799
+001800
+001801
+001806
+001807
+001809
+001810
+001816
+001818
+001821
+001825
+001827
+001828
+001830
+001832
+001833
+001834
+001836
+001837
+001840
+001841
+001842
+001843
+001845
+001847
+001849
+001853
+001854
+001855
+001858
+001860
+001861
+001862
+001864
+001870
+001872
+001875
+001877
+001878
+001881
+001882
+001887
+001888
+001892
+001894
+001896
+001898
+001899
+001901
+001902
+001903
+001904
+001906
+001907
+001911
+001915
+001918
+001920
+001922
+001927
+001928
+001930
+001931
+001932
+001933
+001934
+001936
+001937
+001938
+001940
+001941
+001944
+001945
+001948
+001950
+001952
+001954
+001958
+001960
+001962
+001963
+001964
+001970
+001971
+001972
+001976
+001977
+001978
+001980
+001981
+001982
+001985
+001989
+001995
+001999
+002000
+002001
+002002
+002004
+002006
+002011
+002012
+002015
+002019
+002020
+002021
+002022
+002023
+002024
+002025
+002027
+002030
+002034
+002036
+002037
+002039
+002042
+002043
+002045
+002047
+002049
+002051
+002054
+002055
+002056
+002058
+002061
+002063
+002064
+002067
+002068
+002069
+002070
+002082
+002083
+002086
+002088
+002090
+002091
+002094
+002095
+002096
+002098
+002099
+002101
+002102
+002104
+002108
+002109
+002112
+002114
+002116
+002117
+002120
+002124
+002125
+002126
+002129
+002132
+002134
+002135
+002136
+002139
+002140
+002142
+002145
+002146
+002151
+002152
+002153
+002155
+002156
+002158
+002163
+002165
+002166
+002169
+002170
+002171
+002172
+002174
+002176
+002178
+002179
+002180
+002181
+002182
+002183
+002184
+002186
+002187
+002190
+002191
+002192
+002193
+002194
+002196
+002197
+002199
+002201
+002202
+002208
+002209
+002212
+002213
+002214
+002215
+002218
+002219
+002220
+002221
+002224
+002226
+002228
+002233
+002234
+002237
+002238
+002241
+002244
+002247
+002248
+002249
+002251
+002253
+002255
+002256
+002257
+002259
+002260
+002261
+002263
+002265
+002266
+002267
+002268
+002270
+002272
+002273
+002276
+002277
+002278
+002279
+002280
+002281
+002284
+002285
+002287
+002288
+002290
+002291
+002293
+002300
+002302
+002305
+002306
+002307
+002308
+002310
+002311
+002315
+002318
+002320
+002321
+002323
+002324
+002328
+002329
+002330
+002332
+002333
+002334
+002335
+002337
+002340
+002342
+002343
+002345
+002347
+002348
+002350
+002352
+002354
+002355
+002359
+002361
+002362
+002364
+002366
+002367
+002368
+002369
+002371
+002372
+002373
+002374
+002375
+002376
+002377
+002378
+002382
+002384
+002385
+002387
+002391
+002392
+002393
+002401
+002403
+002404
+002405
+002407
+002410
+002411
+002413
+002415
+002417
+002419
+002420
+002423
+002425
+002427
+002433
+002435
+002436
+002437
+002439
+002441
+002442
+002443
+002444
+002445
+002448
+002450
+002452
+002454
+002456
+002458
+002459
+002460
+002461
+002462
+002465
+002466
+002468
+002470
+002471
+002472
+002476
+002477
+002478
+002479
+002480
+002481
+002483
+002490
+002491
+002492
+002493
+002494
+002496
+002497
+002500
+002501
+002502
+002504
+002505
+002508
+002512
+002513
+002514
+002518
+002519
+002520
+002523
+002524
+002525
+002529
+002533
+002534
+002537
+002539
+002540
+002542
+002544
+002545
+002546
+002547
+002549
+002554
+002555
+002558
+002559
+002561
+002563
+002564
+002565
+002566
+002567
+002569
+002571
+002572
+002578
+002579
+002584
+002585
+002586
+002589
+002590
+002593
+002594
+002595
+002598
+002599
+002600
+002603
+002605
+002606
+002609
+002611
+002613
+002615
+002618
+002621
+002625
+002627
+002632
+002633
+002634
+002635
+002636
+002637
+002641
+002643
+002645
+002646
+002647
+002648
+002649
+002653
+002657
+002658
+002659
+002662
+002664
+002666
+002667
+002668
+002669
+002670
+002675
+002677
+002678
+002680
+002682
+002683
+002684
+002689
+002690
+002691
+002693
+002695
+002696
+002697
+002699
+002702
+002704
+002706
+002709
+002710
+002713
+002714
+002715
+002717
+002718
+002721
+002722
+002723
+002727
+002730
+002732
+002734
+002735
+002737
+002738
+002741
+002744
+002745
+002747
+002749
+002751
+002755
+002757
+002759
+002760
+002762
+002763
+002765
+002766
+002767
+002772
+002774
+002775
+002776
+002778
+002779
+002782
+002783
+002784
+002785
+002786
+002791
+002794
+002795
+002796
+002798
+002800
+002801
+002803
+002804
+002807
+002810
+002812
+002815
+002816
+002817
+002820
+002826
+002827
+002833
+002834
+002835
+002836
+002838
+002841
+002842
+002844
+002845
+002847
+002848
+002854
+002855
+002858
+002859
+002864
+002866
+002867
+002868
+002869
+002870
+002873
+002875
+002879
+002880
+002881
+002884
+002886
+002889
+002891
+002893
+002896
+002899
+002901
+002906
+002910
+002912
+002913
+002914
+002915
+002916
+002917
+002919
+002924
+002931
+002932
+002933
+002934
+002935
+002937
+002938
+002939
+002940
+002941
+002942
+002943
+002944
+002946
+002947
+002952
+002953
+002954
+002956
+002957
+002958
+002960
+002962
+002963
+002965
+002966
+002967
+002969
+002975
+002976
+002977
+002978
+002984
+002986
+002987
+002988
+002989
+002990
+002992
+002994
+002995
+003000
+003002
+003003
+003004
+003005
+003007
+003008
+003009
+003011
+003013
+003015
+003017
+003021
+003023
+003024
+003027
+003028
+003031
+003032
+003034
+003038
+003039
+003042
+003044
+003045
+003047
+003051
+003053
+003054
+003056
+003057
+003058
+003061
+003063
+003064
+003065
+003066
+003072
+003074
+003077
+003078
+003082
+003083
+003085
+003086
+003088
+003089
+003090
+003092
+003093
+003094
+003098
+003100
+003102
+003103
+003105
+003106
+003107
+003108
+003110
+003112
+003116
+003117
+003118
+003120
+003121
+003122
+003124
+003126
+003127
+003129
+003133
+003134
+003135
+003137
+003138
+003140
+003142
+003145
+003146
+003147
+003149
+003150
+003154
+003155
+003157
+003159
+003161
+003162
+003163
+003164
+003165
+003169
+003170
+003175
+003176
+003177
+003178
+003181
+003183
+003184
+003185
+003186
+003188
+003189
+003194
+003195
+003199
+003200
+003202
+003204
+003205
+003207
+003210
+003211
+003213
+003214
+003216
+003218
+003219
+003223
+003228
+003229
+003231
+003233
+003236
+003239
+003240
+003242
+003243
+003244
+003247
+003250
+003253
+003254
+003255
+003256
+003258
+003259
+003260
+003261
+003262
+003269
+003270
+003271
+003272
+003273
+003274
+003279
+003280
+003282
+003284
+003285
+003290
+003292
+003293
+003294
+003296
+003299
+003300
+003301
+003303
+003307
+003308
+003311
+003313
+003316
+003320
+003325
+003327
+003330
+003331
+003335
+003336
+003337
+003338
+003339
+003343
+003344
+003349
+003350
+003351
+003354
+003355
+003356
+003359
+003360
+003362
+003363
+003365
+003367
+003369
+003370
+003373
+003374
+003376
+003377
+003379
+003380
+003382
+003386
+003390
+003391
+003392
+003395
+003396
+003397
+003398
+003401
+003403
+003404
+003406
+003407
+003408
+003410
+003412
+003413
+003415
+003416
+003417
+003419
+003420
+003421
+003422
+003424
+003425
+003429
+003430
+003433
+003435
+003436
+003439
+003441
+003443
+003444
+003449
+003450
+003451
+003452
+003453
+003455
+003458
+003461
+003462
+003464
+003465
+003466
+003468
+003469
+003470
+003477
+003484
+003487
+003489
+003491
+003492
+003493
+003496
+003497
+003499
+003500
+003506
+003508
+003509
+003510
+003511
+003516
+003518
+003519
+003521
+003522
+003524
+003525
+003528
+003529
+003530
+003536
+003537
+003539
+003546
+003548
+003549
+003550
+003551
+003554
+003555
+003556
+003564
+003565
+003566
+003567
+003575
+003576
+003577
+003580
+003585
+003586
+003587
+003588
+003589
+003593
+003594
+003596
+003597
+003599
+003603
+003604
+003605
+003606
+003608
+003609
+003611
+003614
+003618
+003620
+003621
+003622
+003623
+003625
+003627
+003628
+003629
+003632
+003634
+003635
+003636
+003638
+003639
+003640
+003642
+003644
+003645
+003646
+003648
+003651
+003654
+003655
+003656
+003657
+003658
+003660
+003662
+003663
+003664
+003667
+003669
+003671
+003673
+003674
+003675
+003678
+003679
+003681
+003684
+003685
+003688
+003690
+003691
+003694
+003695
+003696
+003698
+003699
+003700
+003703
+003704
+003705
+003706
+003708
+003709
+003711
+003713
+003714
+003717
+003721
+003722
+003727
+003729
+003732
+003735
+003740
+003743
+003748
+003749
+003750
+003751
+003752
+003753
+003754
+003758
+003759
+003760
+003763
+003767
+003772
+003773
+003774
+003779
+003780
+003781
+003783
+003784
+003786
+003788
+003790
+003791
+003792
+003793
+003796
+003797
+003798
+003803
+003806
+003807
+003808
+003809
+003811
+003814
+003817
+003818
+003820
+003821
+003824
+003826
+003827
+003828
+003830
+003834
+003835
+003837
+003838
+003844
+003845
+003846
+003847
+003848
+003849
+003855
+003856
+003857
+003859
+003860
+003861
+003863
+003865
+003866
+003868
+003869
+003871
+003872
+003874
+003876
+003877
+003879
+003885
+003886
+003887
+003889
+003890
+003891
+003895
+003898
+003899
+003905
+003907
+003911
+003912
+003913
+003915
+003918
+003919
+003921
+003923
+003924
+003926
+003932
+003935
+003936
+003937
+003939
+003941
+003945
+003946
+003947
+003948
+003949
+003953
+003954
+003956
+003957
+003960
+003961
+003963
+003965
+003966
+003969
+003970
+003971
+003973
+003974
+003979
+003983
+003984
+003986
+003987
+003988
+003990
+003991
+003992
+003993
+003994
+003996
+003997
+003998
+004003
+004005
+004008
+004009
+004010
+004011
+004012
+004013
+004014
+004015
+004016
+004017
+004019
+004020
+004023
+004025
+004028
+004031
+004033
+004034
+004035
+004037
+004039
+004046
+004047
+004051
+004052
+004057
+004058
+004060
+004066
+004067
+004069
+004073
+004075
+004076
+004077
+004082
+004085
+004087
+004089
+004091
+004092
+004093
+004095
+004100
+004102
+004105
+004106
+004108
+004110
+004111
+004113
+004117
+004120
+004121
+004122
+004129
+004131
+004133
+004135
+004136
+004137
+004138
+004140
+004141
+004142
+004143
+004145
+004146
+004148
+004149
+004150
+004152
+004158
+004163
+004164
+004168
+004169
+004170
+004171
+004174
+004178
+004185
+004186
+004189
+004190
+004191
+004192
+004193
+004194
+004195
+004196
+004200
+004201
+004203
+004204
+004205
+004209
+004212
+004215
+004220
+004221
+004223
+004224
+004228
+004229
+004230
+004231
+004232
+004237
+004239
+004241
+004242
+004244
+004246
+004247
+004253
+004255
+004256
+004257
+004258
+004259
+004263
+004264
+004265
+004269
+004270
+004271
+004272
+004273
+004274
+004275
+004279
+004280
+004281
+004283
+004284
+004286
+004287
+004291
+004292
+004293
+004295
+004296
+004298
+004300
+004303
+004304
+004307
+004310
+004312
+004315
+004318
+004321
+004322
+004323
+004325
+004326
+004327
+004329
+004331
+004333
+004338
+004339
+004341
+004345
+004346
+004347
+004349
+004351
+004352
+004354
+004356
+004359
+004360
+004361
+004364
+004365
+004367
+004368
+004369
+004370
+004371
+004372
+004376
+004379
+004380
+004384
+004386
+004387
+004389
+004390
+004391
+004392
+004396
+004397
+004404
+004405
+004409
+004411
+004421
+004423
+004424
+004429
+004430
+004432
+004433
+004434
+004436
+004437
+004438
+004439
+004441
+004446
+004450
+004452
+004455
+004457
+004459
+004463
+004464
+004466
+004468
+004470
+004471
+004474
+004479
+004481
+004484
+004487
+004488
+004490
+004493
+004494
+004495
+004496
+004498
+004499
+004500
+004502
+004507
+004508
+004509
+004510
+004512
+004514
+004517
+004518
+004519
+004520
+004524
+004526
+004527
+004528
+004530
+004532
+004535
+004537
+004539
+004540
+004542
+004544
+004548
+004549
+004551
+004552
+004553
+004555
+004558
+004562
+004563
+004565
+004566
+004570
+004571
+004574
+004576
+004579
+004581
+004584
+004585
+004587
+004588
+004591
+004592
+004595
+004597
+004600
+004601
+004604
+004605
+004606
+004607
+004609
+004611
+004612
+004618
+004622
+004623
+004625
+004626
+004627
+004628
+004630
+004631
+004632
+004634
+004636
+004643
+004644
+004647
+004648
+004649
+004651
+004652
+004653
+004654
+004655
+004656
+004660
+004662
+004671
+004672
+004673
+004674
+004675
+004676
+004679
+004682
+004683
+004685
+004686
+004687
+004689
+004691
+004692
+004693
+004694
+004699
+004701
+004702
+004705
+004706
+004707
+004708
+004710
+004714
+004715
+004718
+004719
+004722
+004723
+004727
+004732
+004735
+004737
+004742
+004743
+004746
+004747
+004748
+004750
+004753
+004754
+004760
+004761
+004768
+004770
+004773
+004776
+004777
+004779
+004782
+004783
+004785
+004786
+004788
+004789
+004790
+004792
+004793
+004794
+004796
+004797
+004799
+004801
+004805
+004808
+004812
+004814
+004815
+004816
+004818
+004823
+004825
+004826
+004828
+004830
+004831
+004832
+004834
+004836
+004837
+004839
+004840
+004841
+004842
+004846
+004848
+004849
+004850
+004852
+004856
+004857
+004859
+004863
+004866
+004867
+004868
+004869
+004872
+004873
+004876
+004878
+004879
+004882
+004885
+004886
+004890
+004895
+004896
+004897
+004898
+004902
+004903
+004905
+004907
+004910
+004911
+004912
+004913
+004916
+004926
+004928
+004929
+004931
+004935
+004936
+004938
+004939
+004943
+004946
+004948
+004950
+004951
+004953
+004954
+004955
+004956
+004958
+004960
+004961
+004962
+004963
+004966
+004967
+004968
+004972
+004973
+004974
+004976
+004977
+004982
+004983
+004984
+004985
+004986
+004987
+004990
+004991
+004992
+004994
+004995
+004997
+004998
+004999
+005001
+005003
+005004
+005006
+005007
+005014
+005016
+005018
+005020
+005023
+005024
+005026
+005027
+005028
+005029
+005032
+005033
+005036
+005037
+005039
+005042
+005045
+005047
+005052
+005054
+005055
+005056
+005057
+005058
+005061
+005062
+005063
+005064
+005065
+005067
+005068
+005071
+005072
+005073
+005077
+005078
+005079
+005081
+005084
+005085
+005086
+005090
+005093
+005094
+005097
+005101
+005102
+005104
+005107
+005108
+005110
+005111
+005114
+005116
+005121
+005122
+005124
+005128
+005129
+005130
+005131
+005134
+005135
+005136
+005138
+005143
+005144
+005145
+005146
+005150
+005153
+005156
+005159
+005160
+005161
+005168
+005169
+005171
+005173
+005175
+005176
+005177
+005179
+005181
+005183
+005185
+005186
+005189
+005190
+005191
+005195
+005199
+005202
+005203
+005208
+005209
+005210
+005212
+005214
+005215
+005217
+005219
+005220
+005222
+005223
+005224
+005229
+005230
+005231
+005236
+005239
+005242
+005244
+005245
+005246
+005248
+005253
+005254
+005257
+005258
+005259
+005260
+005262
+005263
+005264
+005267
+005268
+005269
+005270
+005273
+005274
+005278
+005281
+005283
+005285
+005288
+005290
+005292
+005293
+005297
+005298
+005303
+005304
+005305
+005306
+005307
+005310
+005311
+005312
+005314
+005315
+005318
+005319
+005320
+005325
+005326
+005327
+005328
+005331
+005336
+005337
+005338
+005340
+005343
+005344
+005345
+005346
+005348
+005349
+005350
+005351
+005352
+005355
+005358
+005360
+005363
+005365
+005367
+005368
+005369
+005370
+005371
+005373
+005374
+005378
+005379
+005380
+005383
+005384
+005385
+005387
+005388
+005389
+005391
+005393
+005395
+005396
+005397
+005398
+005404
+005405
+005406
+005407
+005408
+005410
+005413
+005414
+005416
+005417
+005418
+005419
+005420
+005421
+005423
+005424
+005429
+005430
+005431
+005433
+005434
+005436
+005438
+005439
+005440
+005441
+005445
+005448
+005450
+005451
+005453
+005454
+005455
+005457
+005461
+005465
+005467
+005469
+005470
+005471
+005475
+005478
+005481
+005483
+005485
+005486
+005487
+005489
+005496
+005497
+005499
+005507
+005508
+005509
+005510
+005511
+005514
+005515
+005517
+005518
+005519
+005521
+005522
+005524
+005526
+005527
+005530
+005531
+005535
+005536
+005539
+005541
+005542
+005544
+005547
+005549
+005550
+005552
+005554
+005559
+005563
+005566
+005568
+005573
+005574
+005576
+005577
+005579
+005582
+005583
+005584
+005585
+005586
+005588
+005590
+005591
+005592
+005593
+005599
+005600
+005601
+005603
+005605
+005606
+005608
+005609
+005611
+005613
+005614
+005615
+005618
+005620
+005624
+005625
+005629
+005630
+005631
+005636
+005637
+005639
+005640
+005641
+005644
+005645
+005647
+005648
+005652
+005653
+005654
+005655
+005657
+005658
+005660
+005662
+005664
+005668
+005669
+005672
+005674
+005676
+005679
+005680
+005682
+005685
+005686
+005687
+005693
+005695
+005696
+005697
+005699
+005700
+005701
+005702
+005704
+005705
+005710
+005713
+005714
+005715
+005716
+005718
+005719
+005723
+005728
+005729
+005730
+005731
+005732
+005735
+005736
+005738
+005740
+005741
+005742
+005743
+005747
+005749
+005752
+005755
+005756
+005757
+005760
+005761
+005762
+005764
+005765
+005768
+005769
+005773
+005779
+005780
+005781
+005782
+005783
+005784
+005786
+005788
+005789
+005790
+005791
+005794
+005796
+005799
+005803
+005805
+005806
+005811
+005812
+005813
+005814
+005815
+005817
+005818
+005819
+005821
+005824
+005825
+005826
+005828
+005829
+005830
+005831
+005836
+005838
+005839
+005840
+005841
+005843
+005845
+005850
+005851
+005852
+005853
+005854
+005856
+005859
+005860
+005861
+005863
+005864
+005867
+005868
+005873
+005874
+005875
+005877
+005878
+005879
+005881
+005884
+005885
+005888
+005889
+005893
+005894
+005895
+005897
+005899
+005901
+005903
+005905
+005906
+005908
+005909
+005910
+005911
+005912
+005914
+005917
+005918
+005919
+005920
+005923
+005928
+005930
+005938
+005940
+005947
+005948
+005951
+005952
+005954
+005956
+005960
+005961
+005963
+005964
+005968
+005970
+005971
+005975
+005979
+005980
+005981
+005983
+005984
+005985
+005988
+005989
+005990
+005991
+005992
+005995
+005996
+005998
+006000
+006001
+006004
+006005
+006009
+006011
+006012
+006018
+006020
+006023
+006025
+006026
+006027
+006028
+006029
+006030
+006033
+006035
+006038
+006041
+006042
+006043
+006045
+006046
+006055
+006058
+006061
+006062
+006065
+006066
+006067
+006069
+006070
+006071
+006073
+006074
+006078
+006079
+006084
+006088
+006089
+006091
+006095
+006096
+006097
+006098
+006100
+006103
+006104
+006105
+006107
+006108
+006111
+006117
+006120
+006123
+006124
+006125
+006128
+006129
+006130
+006131
+006133
+006134
+006135
+006136
+006139
+006140
+006141
+006146
+006148
+006150
+006151
+006153
+006156
+006158
+006159
+006161
+006162
+006163
+006166
+006170
+006171
+006172
+006174
+006175
+006176
+006177
+006179
+006180
+006181
+006183
+006184
+006185
+006187
+006188
+006189
+006190
+006196
+006198
+006201
+006202
+006203
+006206
+006208
+006209
+006210
+006212
+006214
+006215
+006216
+006218
+006219
+006220
+006221
+006222
+006223
+006224
+006225
+006229
+006230
+006233
+006234
+006235
+006236
+006238
+006240
+006241
+006243
+006247
+006249
+006250
+006251
+006252
+006254
+006258
+006259
+006260
+006261
+006262
+006264
+006267
+006269
+006270
+006272
+006275
+006276
+006277
+006279
+006281
+006282
+006284
+006285
+006286
+006289
+006290
+006291
+006295
+006296
+006299
+006300
+006301
+006304
+006305
+006306
+006309
+006314
+006318
+006319
+006320
+006321
+006323
+006325
+006329
+006330
+006335
+006337
+006338
+006339
+006341
+006344
+006346
+006348
+006349
+006350
+006351
+006352
+006353
+006355
+006357
+006362
+006363
+006366
+006367
+006369
+006371
+006374
+006375
+006377
+006381
+006382
+006385
+006387
+006391
+006392
+006395
+006396
+006398
+006400
+006404
+006409
+006411
+006417
+006418
+006419
+006421
+006424
+006425
+006427
+006428
+006429
+006430
+006433
+006434
+006436
+006437
+006438
+006440
+006442
+006443
+006444
+006445
+006447
+006448
+006449
+006450
+006455
+006456
+006458
+006459
+006462
+006463
+006465
+006466
+006468
+006470
+006472
+006473
+006474
+006475
+006476
+006480
+006482
+006483
+006484
+006486
+006488
+006492
+006495
+006497
+006499
+006501
+006503
+006506
+006507
+006509
+006512
+006515
+006519
+006520
+006523
+006524
+006529
+006530
+006532
+006534
+006536
+006538
+006542
+006543
+006547
+006548
+006549
+006550
+006551
+006553
+006556
+006560
+006562
+006564
+006565
+006569
+006570
+006572
+006575
+006576
+006578
+006583
+006584
+006585
+006587
+006588
+006593
+006595
+006597
+006599
+006602
+006603
+006605
+006606
+006609
+006610
+006611
+006612
+006617
+006618
+006619
+006621
+006622
+006625
+006626
+006627
+006628
+006631
+006632
+006635
+006636
+006637
+006638
+006643
+006645
+006647
+006648
+006652
+006654
+006657
+006658
+006660
+006661
+006664
+006666
+006667
+006668
+006670
+006671
+006673
+006674
+006677
+006678
+006679
+006681
+006682
+006684
+006687
+006689
+006690
+006694
+006695
+006696
+006697
+006698
+006699
+006702
+006703
+006704
+006706
+006707
+006708
+006709
+006714
+006718
+006719
+006722
+006725
+006726
+006727
+006730
+006731
+006734
+006735
+006736
+006738
+006739
+006740
+006747
+006748
+006751
+006753
+006755
+006759
+006760
+006761
+006762
+006765
+006766
+006768
+006769
+006772
+006773
+006777
+006781
+006782
+006783
+006784
+006786
+006789
+006794
+006797
+006799
+006800
+006802
+006803
+006805
+006806
+006808
+006810
+006813
+006814
+006819
+006821
+006822
+006824
+006825
+006827
+006828
+006829
+006833
+006835
+006836
+006838
+006839
+006840
+006841
+006842
+006844
+006845
+006847
+006848
+006849
+006850
+006852
+006855
+006858
+006859
+006860
+006862
+006864
+006865
+006866
+006867
+006868
+006869
+006874
+006876
+006878
+006880
+006883
+006884
+006886
+006887
+006892
+006893
+006896
+006899
+006900
+006903
+006908
+006909
+006910
+006911
+006912
+006914
+006916
+006917
+006918
+006919
+006922
+006924
+006930
+006931
+006932
+006933
+006934
+006935
+006939
+006940
+006943
+006944
+006945
+006947
+006948
+006949
+006950
+006952
+006953
+006956
+006958
+006959
+006962
+006963
+006965
+006966
+006968
+006971
+006972
+006976
+006981
+006983
+006987
+006988
+006989
+006990
+006994
+006995
+007002
+007003
+007004
+007006
+007007
+007008
+007009
+007011
+007016
+007018
+007020
+007021
+007022
+007023
+007025
+007029
+007031
+007033
+007035
+007036
+007038
+007039
+007040
+007042
+007045
+007046
+007048
+007049
+007050
+007052
+007054
+007056
+007058
+007059
+007062
+007064
+007065
+007068
+007070
+007071
+007072
+007073
+007074
+007075
+007077
+007078
+007079
+007080
+007084
+007086
+007088
+007089
+007090
+007092
+007093
+007095
+007097
+007100
+007101
+007104
+007105
+007108
+007109
+007113
+007114
+007117
+007121
+007122
+007123
+007125
+007128
+007129
+007130
+007132
+007133
+007138
+007139
+007140
+007141
+007144
+007146
+007147
+007148
+007149
+007150
+007152
+007153
+007154
+007159
+007162
+007163
+007165
+007166
+007167
+007168
+007172
+007174
+007177
+007180
+007182
+007184
+007185
+007187
+007189
+007191
+007193
+007194
+007197
+007200
+007204
+007205
+007208
+007210
+007211
+007212
+007213
+007214
+007215
+007216
+007217
+007219
+007222
+007223
+007224
+007227
+007230
+007234
+007236
+007241
+007243
+007244
+007245
+007247
+007249
+007250
+007256
+007258
+007259
+007260
+007261
+007263
+007266
+007270
+007271
+007274
+007275
+007276
+007279
+007280
+007283
+007284
+007285
+007289
+007292
+007294
+007295
+007296
+007297
+007298
+007299
+007300
+007302
+007305
+007308
+007311
+007314
+007318
+007322
+007323
+007325
+007327
+007329
+007330
+007334
+007336
+007343
+007344
+007346
+007350
+007351
+007356
+007359
+007361
+007363
+007365
+007369
+007370
+007372
+007373
+007374
+007375
+007376
+007381
+007383
+007385
+007388
+007389
+007390
+007394
+007396
+007398
+007408
+007410
+007411
+007413
+007414
+007416
+007417
+007419
+007421
+007422
+007424
+007425
+007427
+007431
+007432
+007433
+007435
+007436
+007437
+007438
+007439
+007443
+007445
+007446
+007448
+007449
+007451
+007454
+007457
+007458
+007460
+007461
+007465
+007466
+007467
+007468
+007470
+007474
+007475
+007477
+007479
+007480
+007481
+007482
+007483
+007484
+007486
+007489
+007490
+007491
+007493
+007497
+007498
+007503
+007506
+007511
+007513
+007517
+007519
+007521
+007523
+007524
+007525
+007526
+007527
+007528
+007530
+007533
+007535
+007536
+007537
+007538
+007540
+007543
+007544
+007546
+007547
+007551
+007555
+007558
+007559
+007563
+007565
+007566
+007568
+007570
+007571
+007572
+007575
+007576
+007578
+007579
+007585
+007586
+007590
+007592
+007594
+007600
+007601
+007603
+007605
+007606
+007611
+007612
+007614
+007615
+007618
+007619
+007621
+007622
+007624
+007626
+007629
+007631
+007633
+007637
+007639
+007640
+007642
+007647
+007649
+007650
+007653
+007654
+007655
+007656
+007657
+007662
+007663
+007664
+007666
+007667
+007668
+007670
+007671
+007672
+007673
+007675
+007677
+007678
+007679
+007680
+007682
+007683
+007685
+007687
+007688
+007691
+007692
+007694
+007696
+007697
+007699
+007702
+007704
+007705
+007709
+007712
+007713
+007715
+007718
+007720
+007721
+007723
+007724
+007727
+007729
+007731
+007732
+007735
+007736
+007740
+007742
+007743
+007745
+007746
+007748
+007749
+007751
+007753
+007754
+007758
+007760
+007762
+007763
+007765
+007767
+007768
+007772
+007773
+007775
+007776
+007777
+007779
+007781
+007786
+007790
+007791
+007793
+007795
+007798
+007799
+007803
+007809
+007810
+007812
+007813
+007814
+007815
+007819
+007820
+007821
+007824
+007826
+007831
+007833
+007834
+007836
+007838
+007840
+007841
+007843
+007845
+007847
+007853
+007854
+007855
+007856
+007857
+007859
+007863
+007864
+007865
+007868
+007869
+007872
+007873
+007876
+007877
+007878
+007883
+007884
+007885
+007886
+007889
+007890
+007897
+007898
+007899
+007900
+007901
+007902
+007905
+007908
+007909
+007910
+007911
+007914
+007915
+007916
+007919
+007920
+007921
+007923
+007924
+007925
+007926
+007928
+007931
+007932
+007933
+007935
+007939
+007940
+007943
+007946
+007947
+007950
+007953
+007954
+007956
+007958
+007959
+007963
+007964
+007968
+007970
+007971
+007974
+007976
+007979
+007980
+007984
+007987
+007991
+007996
+007997
+007998
+007999
+008001
+008002
+008004
+008005
+008008
+008009
+008012
+008017
+008019
+008023
+008024
+008026
+008029
+008031
+008032
+008033
+008036
+008037
+008040
+008042
+008043
+008044
+008048
+008049
+008051
+008053
+008057
+008060
+008061
+008062
+008063
+008064
+008067
+008068
+008069
+008072
+008075
+008076
+008079
+008082
+008083
+008084
+008085
+008086
+008087
+008091
+008093
+008095
+008096
+008098
+008100
+008101
+008103
+008105
+008106
+008107
+008108
+008112
+008115
+008116
+008117
+008121
+008122
+008125
+008127
+008130
+008132
+008137
+008138
+008139
+008140
+008141
+008142
+008144
+008150
+008151
+008159
+008160
+008163
+008164
+008166
+008168
+008169
+008171
+008173
+008174
+008175
+008177
+008180
+008186
+008188
+008189
+008190
+008191
+008197
+008199
+008200
+008202
+008203
+008204
+008208
+008209
+008211
+008213
+008216
+008218
+008220
+008222
+008223
+008224
+008225
+008226
+008229
+008232
+008235
+008236
+008241
+008244
+008248
+008250
+008251
+008252
+008253
+008254
+008258
+008260
+008261
+008262
+008263
+008268
+008269
+008272
+008275
+008279
+008280
+008281
+008282
+008284
+008285
+008292
+008293
+008294
+008295
+008296
+008297
+008299
+008300
+008301
+008302
+008306
+008307
+008310
+008311
+008312
+008313
+008315
+008316
+008317
+008318
+008319
+008320
+008322
+008323
+008326
+008327
+008329
+008332
+008335
+008336
+008338
+008341
+008342
+008345
+008346
+008349
+008351
+008355
+008359
+008360
+008364
+008365
+008368
+008370
+008372
+008374
+008376
+008381
+008384
+008385
+008386
+008387
+008388
+008390
+008391
+008397
+008398
+008403
+008409
+008410
+008413
+008415
+008416
+008422
+008423
+008424
+008425
+008426
+008427
+008429
+008430
+008433
+008434
+008437
+008438
+008442
+008443
+008444
+008445
+008449
+008450
+008452
+008453
+008454
+008456
+008461
+008462
+008465
+008466
+008467
+008468
+008470
+008472
+008475
+008477
+008478
+008482
+008483
+008484
+008485
+008492
+008494
+008495
+008498
+008499
+008502
+008503
+008506
+008509
+008512
+008513
+008514
+008517
+008518
+008519
+008521
+008522
+008523
+008524
+008526
+008529
+008530
+008533
+008534
+008535
+008536
+008541
+008542
+008549
+008550
+008553
+008556
+008557
+008558
+008559
+008562
+008564
+008568
+008572
+008573
+008576
+008581
+008582
+008584
+008585
+008586
+008587
+008588
+008592
+008595
+008596
+008601
+008602
+008604
+008606
+008607
+008608
+008610
+008612
+008615
+008617
+008618
+008620
+008621
+008624
+008628
+008633
+008635
+008636
+008638
+008639
+008644
+008645
+008647
+008653
+008654
+008655
+008663
+008665
+008667
+008670
+008676
+008680
+008683
+008687
+008688
+008690
+008691
+008692
+008695
+008698
+008699
+008701
+008702
+008706
+008709
+008710
+008713
+008716
+008717
+008718
+008720
+008722
+008723
+008725
+008727
+008728
+008730
+008731
+008732
+008733
+008738
+008739
+008741
+008742
+008744
+008747
+008748
+008749
+008750
+008752
+008753
+008755
+008756
+008757
+008759
+008760
+008764
+008766
+008768
+008769
+008770
+008771
+008772
+008773
+008775
+008776
+008783
+008784
+008790
+008793
+008794
+008796
+008799
+008801
+008805
+008806
+008809
+008810
+008811
+008813
+008814
+008815
+008817
+008819
+008822
+008823
+008826
+008831
+008833
+008835
+008836
+008837
+008838
+008840
+008841
+008843
+008847
+008848
+008849
+008854
+008856
+008858
+008859
+008862
+008865
+008867
+008871
+008872
+008873
+008874
+008876
+008878
+008879
+008880
+008883
+008884
+008885
+008886
+008888
+008890
+008891
+008892
+008900
+008905
+008909
+008911
+008913
+008914
+008917
+008919
+008920
+008921
+008923
+008926
+008927
+008929
+008930
+008931
+008932
+008933
+008936
+008939
+008940
+008942
+008943
+008944
+008948
+008951
+008953
+008955
+008958
+008960
+008961
+008962
+008965
+008966
+008967
+008968
+008969
+008970
+008971
+008973
+008975
+008976
+008978
+008979
+008980
+008982
+008983
+008985
+008987
+008988
+008989
+008995
+008997
+008999
+009000
+009002
+009004
+009005
+009006
+009007
+009015
+009016
+009018
+009019
+009020
+009022
+009024
+009027
+009029
+009032
+009034
+009035
+009036
+009037
+009039
+009042
+009045
+009048
+009049
+009051
+009053
+009058
+009059
+009060
+009063
+009064
+009066
+009068
+009072
+009073
+009078
+009079
+009080
+009085
+009086
+009087
+009089
+009091
+009094
+009098
+009099
+009100
+009105
+009106
+009108
+009112
+009113
+009114
+009116
+009117
+009121
+009123
+009126
+009128
+009129
+009131
+009133
+009136
+009138
+009141
+009144
+009147
+009148
+009150
+009151
+009153
+009155
+009157
+009159
+009160
+009161
+009162
+009163
+009166
+009168
+009173
+009174
+009175
+009177
+009178
+009179
+009180
+009181
+009184
+009185
+009186
+009187
+009189
+009191
+009192
+009193
+009194
+009195
+009196
+009197
+009200
+009202
+009205
+009208
+009209
+009212
+009213
+009214
+009215
+009218
+009221
+009224
+009227
+009230
+009236
+009238
+009239
+009242
+009244
+009245
+009246
+009247
+009249
+009250
+009251
+009252
+009254
+009255
+009259
+009268
+009269
+009270
+009271
+009272
+009273
+009278
+009279
+009281
+009282
+009283
+009285
+009286
+009287
+009288
+009289
+009290
+009291
+009295
+009296
+009299
+009303
+009306
+009307
+009308
+009309
+009312
+009315
+009316
+009318
+009323
+009324
+009325
+009326
+009327
+009330
+009331
+009333
+009334
+009336
+009337
+009339
+009342
+009343
+009347
+009348
+009349
+009350
+009351
+009354
+009358
+009359
+009362
+009365
+009368
+009371
+009373
+009374
+009375
+009377
+009378
+009382
+009386
+009388
+009389
+009392
+009393
+009394
+009398
+009401
+009405
+009406
+009407
+009408
+009409
+009410
+009411
+009412
+009413
+009414
+009417
+009418
+009419
+009420
+009421
+009422
+009424
+009429
+009432
+009433
+009434
+009437
+009438
+009439
+009440
+009443
+009445
+009446
+009448
+009454
+009455
+009456
+009457
+009458
+009459
+009460
+009461
+009463
+009464
+009465
+009466
+009468
+009469
+009470
+009472
+009476
+009477
+009479
+009480
+009481
+009484
+009488
+009490
+009491
+009494
+009496
+009497
+009499
+009500
+009502
+009504
+009507
+009508
+009512
+009515
+009516
+009517
+009518
+009519
+009520
+009523
+009524
+009526
+009527
+009528
+009531
+009532
+009533
+009537
+009540
+009541
+009542
+009543
+009545
+009546
+009549
+009550
+009551
+009557
+009558
+009560
+009562
+009565
+009566
+009567
+009568
+009571
+009573
+009576
+009577
+009579
+009580
+009584
+009585
+009586
+009587
+009588
+009591
+009596
+009597
+009598
+009600
+009603
+009605
+009609
+009611
+009613
+009614
+009615
+009617
+009618
+009619
+009620
+009621
+009623
+009627
+009629
+009634
+009636
+009637
+009638
+009641
+009644
+009647
+009649
+009650
+009654
+009655
+009656
+009658
+009659
+009664
+009666
+009667
+009668
+009670
+009671
+009676
+009678
+009679
+009681
+009684
+009685
+009686
+009687
+009691
+009692
+009693
+009695
+009698
+009699
+009700
+009702
+009703
+009706
+009707
+009709
+009710
+009711
+009712
+009713
+009717
+009718
+009719
+009721
+009724
+009726
+009729
+009732
+009733
+009734
+009735
+009737
+009738
+009743
+009745
+009746
+009747
+009748
+009749
+009754
+009755
+009756
+009758
+009761
+009762
+009763
+009764
+009767
+009772
+009773
+009774
+009776
+009778
+009780
+009781
+009785
+009789
+009790
+009792
+009794
+009796
+009797
+009800
+009801
+009805
+009807
+009808
+009809
+009810
+009813
+009816
+009819
+009822
+009823
+009825
+009828
+009830
+009831
+009832
+009833
+009834
+009836
+009839
+009841
+009842
+009845
+009848
+009851
+009852
+009855
+009858
+009859
+009860
+009862
+009863
+009865
+009867
+009868
+009869
+009870
+009872
+009874
+009877
+009878
+009879
+009880
+009881
+009882
+009884
+009886
+009887
+009894
+009896
+009897
+009898
+009900
+009902
+009904
+009905
+009908
+009911
+009913
+009917
+009918
+009920
+009923
+009926
+009932
+009935
+009938
+009939
+009940
+009942
+009944
+009945
+009946
+009947
+009949
+009950
+009954
+009955
+009958
+009959
+009961
+2008_000002
+2008_000003
+2008_000007
+2008_000008
+2008_000009
+2008_000015
+2008_000016
+2008_000019
+2008_000021
+2008_000023
+2008_000026
+2008_000027
+2008_000028
+2008_000032
+2008_000033
+2008_000034
+2008_000036
+2008_000037
+2008_000041
+2008_000042
+2008_000043
+2008_000045
+2008_000050
+2008_000051
+2008_000052
+2008_000053
+2008_000054
+2008_000056
+2008_000059
+2008_000060
+2008_000062
+2008_000064
+2008_000066
+2008_000067
+2008_000070
+2008_000073
+2008_000074
+2008_000075
+2008_000076
+2008_000078
+2008_000080
+2008_000082
+2008_000084
+2008_000085
+2008_000089
+2008_000090
+2008_000093
+2008_000095
+2008_000096
+2008_000097
+2008_000099
+2008_000103
+2008_000105
+2008_000107
+2008_000109
+2008_000112
+2008_000115
+2008_000116
+2008_000119
+2008_000120
+2008_000123
+2008_000128
+2008_000131
+2008_000132
+2008_000133
+2008_000134
+2008_000138
+2008_000140
+2008_000141
+2008_000142
+2008_000143
+2008_000144
+2008_000145
+2008_000148
+2008_000149
+2008_000151
+2008_000154
+2008_000162
+2008_000163
+2008_000174
+2008_000176
+2008_000177
+2008_000181
+2008_000182
+2008_000183
+2008_000185
+2008_000187
+2008_000188
+2008_000189
+2008_000190
+2008_000191
+2008_000192
+2008_000193
+2008_000194
+2008_000195
+2008_000196
+2008_000197
+2008_000199
+2008_000202
+2008_000203
+2008_000204
+2008_000207
+2008_000213
+2008_000215
+2008_000217
+2008_000219
+2008_000222
+2008_000223
+2008_000226
+2008_000227
+2008_000233
+2008_000234
+2008_000235
+2008_000236
+2008_000237
+2008_000238
+2008_000239
+2008_000243
+2008_000244
+2008_000246
+2008_000251
+2008_000252
+2008_000253
+2008_000254
+2008_000255
+2008_000257
+2008_000259
+2008_000260
+2008_000261
+2008_000262
+2008_000264
+2008_000266
+2008_000268
+2008_000270
+2008_000271
+2008_000272
+2008_000273
+2008_000274
+2008_000275
+2008_000277
+2008_000278
+2008_000281
+2008_000283
+2008_000284
+2008_000287
+2008_000289
+2008_000290
+2008_000291
+2008_000297
+2008_000298
+2008_000304
+2008_000305
+2008_000306
+2008_000307
+2008_000309
+2008_000311
+2008_000313
+2008_000315
+2008_000316
+2008_000318
+2008_000321
+2008_000328
+2008_000330
+2008_000335
+2008_000336
+2008_000338
+2008_000339
+2008_000340
+2008_000342
+2008_000343
+2008_000345
+2008_000346
+2008_000348
+2008_000350
+2008_000354
+2008_000356
+2008_000358
+2008_000359
+2008_000361
+2008_000364
+2008_000365
+2008_000367
+2008_000371
+2008_000373
+2008_000376
+2008_000378
+2008_000380
+2008_000381
+2008_000382
+2008_000383
+2008_000391
+2008_000392
+2008_000393
+2008_000397
+2008_000398
+2008_000399
+2008_000400
+2008_000401
+2008_000403
+2008_000405
+2008_000406
+2008_000407
+2008_000408
+2008_000413
+2008_000414
+2008_000415
+2008_000416
+2008_000418
+2008_000419
+2008_000421
+2008_000422
+2008_000423
+2008_000424
+2008_000426
+2008_000428
+2008_000432
+2008_000435
+2008_000436
+2008_000437
+2008_000442
+2008_000443
+2008_000445
+2008_000446
+2008_000447
+2008_000448
+2008_000452
+2008_000455
+2008_000457
+2008_000461
+2008_000464
+2008_000465
+2008_000466
+2008_000469
+2008_000470
+2008_000471
+2008_000472
+2008_000473
+2008_000474
+2008_000475
+2008_000480
+2008_000481
+2008_000488
+2008_000489
+2008_000491
+2008_000492
+2008_000493
+2008_000495
+2008_000496
+2008_000498
+2008_000499
+2008_000501
+2008_000502
+2008_000505
+2008_000510
+2008_000511
+2008_000512
+2008_000514
+2008_000515
+2008_000516
+2008_000519
+2008_000522
+2008_000527
+2008_000531
+2008_000532
+2008_000533
+2008_000535
+2008_000536
+2008_000540
+2008_000541
+2008_000544
+2008_000545
+2008_000547
+2008_000548
+2008_000552
+2008_000553
+2008_000558
+2008_000559
+2008_000561
+2008_000562
+2008_000563
+2008_000564
+2008_000566
+2008_000567
+2008_000568
+2008_000569
+2008_000572
+2008_000573
+2008_000578
+2008_000579
+2008_000581
+2008_000583
+2008_000584
+2008_000585
+2008_000588
+2008_000589
+2008_000595
+2008_000599
+2008_000602
+2008_000605
+2008_000607
+2008_000609
+2008_000613
+2008_000614
+2008_000615
+2008_000619
+2008_000620
+2008_000622
+2008_000623
+2008_000626
+2008_000628
+2008_000629
+2008_000630
+2008_000634
+2008_000636
+2008_000640
+2008_000641
+2008_000645
+2008_000646
+2008_000647
+2008_000648
+2008_000650
+2008_000652
+2008_000655
+2008_000656
+2008_000657
+2008_000659
+2008_000660
+2008_000661
+2008_000662
+2008_000666
+2008_000669
+2008_000670
+2008_000672
+2008_000673
+2008_000674
+2008_000676
+2008_000677
+2008_000678
+2008_000683
+2008_000689
+2008_000690
+2008_000691
+2008_000694
+2008_000695
+2008_000696
+2008_000697
+2008_000699
+2008_000700
+2008_000703
+2008_000704
+2008_000705
+2008_000706
+2008_000711
+2008_000714
+2008_000716
+2008_000719
+2008_000721
+2008_000723
+2008_000724
+2008_000725
+2008_000726
+2008_000727
+2008_000729
+2008_000731
+2008_000732
+2008_000733
+2008_000734
+2008_000737
+2008_000740
+2008_000742
+2008_000745
+2008_000748
+2008_000753
+2008_000756
+2008_000758
+2008_000760
+2008_000761
+2008_000764
+2008_000765
+2008_000769
+2008_000775
+2008_000776
+2008_000777
+2008_000778
+2008_000780
+2008_000782
+2008_000783
+2008_000785
+2008_000787
+2008_000788
+2008_000790
+2008_000792
+2008_000793
+2008_000795
+2008_000796
+2008_000798
+2008_000801
+2008_000803
+2008_000804
+2008_000805
+2008_000806
+2008_000808
+2008_000811
+2008_000814
+2008_000815
+2008_000817
+2008_000824
+2008_000825
+2008_000828
+2008_000829
+2008_000832
+2008_000833
+2008_000834
+2008_000835
+2008_000837
+2008_000839
+2008_000841
+2008_000842
+2008_000844
+2008_000847
+2008_000848
+2008_000851
+2008_000853
+2008_000854
+2008_000857
+2008_000858
+2008_000860
+2008_000861
+2008_000863
+2008_000864
+2008_000867
+2008_000868
+2008_000870
+2008_000873
+2008_000875
+2008_000876
+2008_000878
+2008_000880
+2008_000881
+2008_000883
+2008_000884
+2008_000885
+2008_000887
+2008_000897
+2008_000899
+2008_000901
+2008_000902
+2008_000904
+2008_000905
+2008_000908
+2008_000910
+2008_000911
+2008_000912
+2008_000914
+2008_000915
+2008_000916
+2008_000917
+2008_000919
+2008_000922
+2008_000923
+2008_000924
+2008_000928
+2008_000931
+2008_000934
+2008_000936
+2008_000939
+2008_000940
+2008_000941
+2008_000942
+2008_000943
+2008_000944
+2008_000950
+2008_000952
+2008_000953
+2008_000956
+2008_000957
+2008_000959
+2008_000960
+2008_000964
+2008_000965
+2008_000970
+2008_000971
+2008_000972
+2008_000973
+2008_000976
+2008_000979
+2008_000981
+2008_000982
+2008_000984
+2008_000985
+2008_000987
+2008_000992
+2008_000993
+2008_000999
+2008_001004
+2008_001007
+2008_001009
+2008_001012
+2008_001013
+2008_001018
+2008_001020
+2008_001021
+2008_001022
+2008_001023
+2008_001024
+2008_001026
+2008_001028
+2008_001030
+2008_001031
+2008_001034
+2008_001035
+2008_001036
+2008_001039
+2008_001040
+2008_001041
+2008_001042
+2008_001046
+2008_001047
+2008_001048
+2008_001052
+2008_001054
+2008_001055
+2008_001056
+2008_001057
+2008_001060
+2008_001062
+2008_001063
+2008_001066
+2008_001068
+2008_001070
+2008_001071
+2008_001073
+2008_001074
+2008_001075
+2008_001076
+2008_001077
+2008_001078
+2008_001080
+2008_001081
+2008_001083
+2008_001089
+2008_001090
+2008_001092
+2008_001098
+2008_001099
+2008_001104
+2008_001105
+2008_001106
+2008_001111
+2008_001112
+2008_001113
+2008_001114
+2008_001115
+2008_001118
+2008_001119
+2008_001120
+2008_001121
+2008_001122
+2008_001130
+2008_001133
+2008_001134
+2008_001135
+2008_001136
+2008_001137
+2008_001139
+2008_001140
+2008_001142
+2008_001143
+2008_001147
+2008_001150
+2008_001154
+2008_001155
+2008_001158
+2008_001159
+2008_001160
+2008_001161
+2008_001164
+2008_001166
+2008_001167
+2008_001168
+2008_001169
+2008_001170
+2008_001171
+2008_001177
+2008_001182
+2008_001183
+2008_001185
+2008_001188
+2008_001189
+2008_001190
+2008_001192
+2008_001194
+2008_001196
+2008_001199
+2008_001202
+2008_001203
+2008_001205
+2008_001206
+2008_001208
+2008_001210
+2008_001215
+2008_001218
+2008_001219
+2008_001220
+2008_001221
+2008_001223
+2008_001225
+2008_001226
+2008_001227
+2008_001230
+2008_001231
+2008_001235
+2008_001236
+2008_001238
+2008_001241
+2008_001245
+2008_001248
+2008_001249
+2008_001255
+2008_001257
+2008_001260
+2008_001262
+2008_001263
+2008_001264
+2008_001267
+2008_001271
+2008_001272
+2008_001274
+2008_001275
+2008_001278
+2008_001283
+2008_001284
+2008_001285
+2008_001290
+2008_001294
+2008_001296
+2008_001299
+2008_001301
+2008_001302
+2008_001304
+2008_001306
+2008_001307
+2008_001308
+2008_001310
+2008_001312
+2008_001314
+2008_001318
+2008_001320
+2008_001322
+2008_001325
+2008_001329
+2008_001333
+2008_001334
+2008_001335
+2008_001336
+2008_001338
+2008_001340
+2008_001344
+2008_001346
+2008_001349
+2008_001350
+2008_001351
+2008_001353
+2008_001356
+2008_001357
+2008_001358
+2008_001359
+2008_001366
+2008_001367
+2008_001369
+2008_001373
+2008_001374
+2008_001375
+2008_001376
+2008_001379
+2008_001380
+2008_001382
+2008_001383
+2008_001385
+2008_001387
+2008_001388
+2008_001389
+2008_001390
+2008_001391
+2008_001395
+2008_001399
+2008_001401
+2008_001402
+2008_001404
+2008_001405
+2008_001406
+2008_001408
+2008_001410
+2008_001413
+2008_001414
+2008_001415
+2008_001419
+2008_001420
+2008_001427
+2008_001428
+2008_001429
+2008_001430
+2008_001431
+2008_001432
+2008_001433
+2008_001434
+2008_001436
+2008_001437
+2008_001439
+2008_001440
+2008_001444
+2008_001445
+2008_001446
+2008_001448
+2008_001451
+2008_001454
+2008_001455
+2008_001456
+2008_001460
+2008_001461
+2008_001462
+2008_001464
+2008_001466
+2008_001467
+2008_001468
+2008_001470
+2008_001475
+2008_001478
+2008_001479
+2008_001481
+2008_001482
+2008_001486
+2008_001488
+2008_001491
+2008_001493
+2008_001494
+2008_001495
+2008_001498
+2008_001500
+2008_001501
+2008_001503
+2008_001504
+2008_001510
+2008_001513
+2008_001514
+2008_001516
+2008_001520
+2008_001522
+2008_001523
+2008_001525
+2008_001527
+2008_001529
+2008_001531
+2008_001533
+2008_001534
+2008_001536
+2008_001538
+2008_001539
+2008_001540
+2008_001541
+2008_001542
+2008_001543
+2008_001544
+2008_001546
+2008_001547
+2008_001549
+2008_001550
+2008_001551
+2008_001553
+2008_001563
+2008_001564
+2008_001566
+2008_001574
+2008_001575
+2008_001576
+2008_001577
+2008_001580
+2008_001582
+2008_001586
+2008_001589
+2008_001590
+2008_001591
+2008_001592
+2008_001593
+2008_001594
+2008_001596
+2008_001598
+2008_001601
+2008_001602
+2008_001605
+2008_001607
+2008_001609
+2008_001610
+2008_001613
+2008_001615
+2008_001617
+2008_001619
+2008_001620
+2008_001622
+2008_001624
+2008_001625
+2008_001626
+2008_001629
+2008_001631
+2008_001632
+2008_001636
+2008_001638
+2008_001640
+2008_001641
+2008_001643
+2008_001645
+2008_001648
+2008_001649
+2008_001652
+2008_001653
+2008_001655
+2008_001659
+2008_001660
+2008_001661
+2008_001663
+2008_001666
+2008_001667
+2008_001668
+2008_001669
+2008_001670
+2008_001673
+2008_001676
+2008_001679
+2008_001680
+2008_001681
+2008_001682
+2008_001688
+2008_001690
+2008_001691
+2008_001692
+2008_001694
+2008_001697
+2008_001699
+2008_001702
+2008_001704
+2008_001706
+2008_001708
+2008_001709
+2008_001710
+2008_001712
+2008_001714
+2008_001715
+2008_001716
+2008_001717
+2008_001719
+2008_001722
+2008_001723
+2008_001724
+2008_001727
+2008_001729
+2008_001730
+2008_001731
+2008_001735
+2008_001736
+2008_001737
+2008_001741
+2008_001742
+2008_001744
+2008_001745
+2008_001746
+2008_001750
+2008_001751
+2008_001757
+2008_001758
+2008_001761
+2008_001763
+2008_001764
+2008_001765
+2008_001769
+2008_001770
+2008_001772
+2008_001773
+2008_001774
+2008_001775
+2008_001781
+2008_001782
+2008_001783
+2008_001784
+2008_001787
+2008_001789
+2008_001791
+2008_001792
+2008_001796
+2008_001797
+2008_001799
+2008_001801
+2008_001802
+2008_001805
+2008_001806
+2008_001808
+2008_001809
+2008_001810
+2008_001811
+2008_001812
+2008_001813
+2008_001814
+2008_001815
+2008_001816
+2008_001820
+2008_001821
+2008_001823
+2008_001825
+2008_001829
+2008_001830
+2008_001832
+2008_001834
+2008_001836
+2008_001837
+2008_001838
+2008_001841
+2008_001842
+2008_001843
+2008_001845
+2008_001849
+2008_001850
+2008_001852
+2008_001854
+2008_001856
+2008_001858
+2008_001860
+2008_001862
+2008_001863
+2008_001865
+2008_001866
+2008_001867
+2008_001869
+2008_001871
+2008_001872
+2008_001874
+2008_001876
+2008_001880
+2008_001881
+2008_001882
+2008_001885
+2008_001888
+2008_001894
+2008_001895
+2008_001896
+2008_001899
+2008_001903
+2008_001905
+2008_001907
+2008_001908
+2008_001909
+2008_001910
+2008_001911
+2008_001914
+2008_001919
+2008_001920
+2008_001921
+2008_001926
+2008_001928
+2008_001929
+2008_001930
+2008_001932
+2008_001934
+2008_001937
+2008_001941
+2008_001945
+2008_001946
+2008_001947
+2008_001951
+2008_001955
+2008_001956
+2008_001957
+2008_001958
+2008_001961
+2008_001965
+2008_001966
+2008_001967
+2008_001969
+2008_001970
+2008_001971
+2008_001977
+2008_001978
+2008_001979
+2008_001980
+2008_001982
+2008_001985
+2008_001986
+2008_001987
+2008_001989
+2008_001992
+2008_001997
+2008_001998
+2008_002000
+2008_002001
+2008_002002
+2008_002003
+2008_002004
+2008_002005
+2008_002007
+2008_002009
+2008_002011
+2008_002013
+2008_002017
+2008_002021
+2008_002023
+2008_002026
+2008_002031
+2008_002032
+2008_002033
+2008_002035
+2008_002036
+2008_002037
+2008_002039
+2008_002042
+2008_002043
+2008_002045
+2008_002046
+2008_002047
+2008_002052
+2008_002056
+2008_002058
+2008_002061
+2008_002062
+2008_002064
+2008_002066
+2008_002067
+2008_002069
+2008_002071
+2008_002073
+2008_002079
+2008_002080
+2008_002082
+2008_002084
+2008_002086
+2008_002088
+2008_002092
+2008_002093
+2008_002094
+2008_002096
+2008_002098
+2008_002099
+2008_002103
+2008_002107
+2008_002112
+2008_002113
+2008_002114
+2008_002115
+2008_002116
+2008_002117
+2008_002118
+2008_002119
+2008_002123
+2008_002124
+2008_002129
+2008_002131
+2008_002132
+2008_002138
+2008_002140
+2008_002144
+2008_002145
+2008_002146
+2008_002148
+2008_002150
+2008_002151
+2008_002152
+2008_002153
+2008_002155
+2008_002156
+2008_002158
+2008_002160
+2008_002162
+2008_002167
+2008_002169
+2008_002172
+2008_002175
+2008_002176
+2008_002177
+2008_002179
+2008_002181
+2008_002182
+2008_002185
+2008_002191
+2008_002193
+2008_002194
+2008_002195
+2008_002197
+2008_002198
+2008_002199
+2008_002200
+2008_002201
+2008_002202
+2008_002204
+2008_002205
+2008_002206
+2008_002207
+2008_002208
+2008_002209
+2008_002210
+2008_002212
+2008_002215
+2008_002218
+2008_002220
+2008_002221
+2008_002222
+2008_002223
+2008_002225
+2008_002227
+2008_002229
+2008_002231
+2008_002234
+2008_002236
+2008_002239
+2008_002240
+2008_002241
+2008_002243
+2008_002244
+2008_002247
+2008_002248
+2008_002250
+2008_002251
+2008_002255
+2008_002258
+2008_002259
+2008_002262
+2008_002267
+2008_002269
+2008_002270
+2008_002272
+2008_002273
+2008_002278
+2008_002279
+2008_002280
+2008_002281
+2008_002283
+2008_002288
+2008_002292
+2008_002293
+2008_002294
+2008_002296
+2008_002298
+2008_002299
+2008_002304
+2008_002305
+2008_002307
+2008_002311
+2008_002312
+2008_002314
+2008_002317
+2008_002321
+2008_002322
+2008_002324
+2008_002325
+2008_002327
+2008_002328
+2008_002329
+2008_002330
+2008_002331
+2008_002335
+2008_002338
+2008_002340
+2008_002343
+2008_002344
+2008_002347
+2008_002349
+2008_002350
+2008_002356
+2008_002357
+2008_002358
+2008_002359
+2008_002361
+2008_002362
+2008_002365
+2008_002366
+2008_002368
+2008_002369
+2008_002370
+2008_002372
+2008_002374
+2008_002377
+2008_002378
+2008_002379
+2008_002383
+2008_002384
+2008_002389
+2008_002395
+2008_002399
+2008_002401
+2008_002403
+2008_002404
+2008_002405
+2008_002408
+2008_002410
+2008_002411
+2008_002412
+2008_002414
+2008_002418
+2008_002419
+2008_002422
+2008_002424
+2008_002425
+2008_002428
+2008_002429
+2008_002430
+2008_002434
+2008_002436
+2008_002437
+2008_002438
+2008_002439
+2008_002441
+2008_002442
+2008_002444
+2008_002445
+2008_002446
+2008_002448
+2008_002451
+2008_002452
+2008_002454
+2008_002456
+2008_002457
+2008_002458
+2008_002459
+2008_002461
+2008_002464
+2008_002465
+2008_002466
+2008_002467
+2008_002470
+2008_002471
+2008_002473
+2008_002477
+2008_002481
+2008_002482
+2008_002483
+2008_002484
+2008_002485
+2008_002487
+2008_002491
+2008_002492
+2008_002494
+2008_002495
+2008_002499
+2008_002501
+2008_002502
+2008_002504
+2008_002506
+2008_002508
+2008_002509
+2008_002510
+2008_002512
+2008_002514
+2008_002515
+2008_002516
+2008_002521
+2008_002523
+2008_002524
+2008_002526
+2008_002527
+2008_002533
+2008_002536
+2008_002540
+2008_002541
+2008_002542
+2008_002543
+2008_002547
+2008_002549
+2008_002551
+2008_002555
+2008_002558
+2008_002562
+2008_002564
+2008_002566
+2008_002567
+2008_002568
+2008_002574
+2008_002575
+2008_002576
+2008_002578
+2008_002579
+2008_002583
+2008_002584
+2008_002588
+2008_002589
+2008_002590
+2008_002597
+2008_002598
+2008_002599
+2008_002601
+2008_002603
+2008_002606
+2008_002610
+2008_002612
+2008_002613
+2008_002616
+2008_002621
+2008_002622
+2008_002623
+2008_002624
+2008_002625
+2008_002631
+2008_002634
+2008_002638
+2008_002639
+2008_002640
+2008_002641
+2008_002643
+2008_002645
+2008_002647
+2008_002648
+2008_002649
+2008_002650
+2008_002652
+2008_002653
+2008_002662
+2008_002665
+2008_002666
+2008_002668
+2008_002670
+2008_002672
+2008_002673
+2008_002674
+2008_002675
+2008_002676
+2008_002677
+2008_002678
+2008_002679
+2008_002680
+2008_002681
+2008_002682
+2008_002684
+2008_002686
+2008_002687
+2008_002696
+2008_002697
+2008_002698
+2008_002700
+2008_002701
+2008_002704
+2008_002705
+2008_002709
+2008_002710
+2008_002712
+2008_002714
+2008_002715
+2008_002716
+2008_002718
+2008_002719
+2008_002720
+2008_002725
+2008_002728
+2008_002730
+2008_002732
+2008_002733
+2008_002735
+2008_002736
+2008_002738
+2008_002741
+2008_002746
+2008_002749
+2008_002750
+2008_002751
+2008_002752
+2008_002753
+2008_002756
+2008_002758
+2008_002760
+2008_002762
+2008_002766
+2008_002767
+2008_002768
+2008_002772
+2008_002773
+2008_002774
+2008_002775
+2008_002776
+2008_002778
+2008_002783
+2008_002784
+2008_002787
+2008_002789
+2008_002791
+2008_002792
+2008_002793
+2008_002794
+2008_002795
+2008_002801
+2008_002804
+2008_002806
+2008_002808
+2008_002809
+2008_002811
+2008_002813
+2008_002814
+2008_002817
+2008_002820
+2008_002823
+2008_002826
+2008_002829
+2008_002830
+2008_002831
+2008_002834
+2008_002835
+2008_002838
+2008_002842
+2008_002843
+2008_002845
+2008_002847
+2008_002848
+2008_002850
+2008_002852
+2008_002854
+2008_002856
+2008_002857
+2008_002859
+2008_002860
+2008_002864
+2008_002866
+2008_002868
+2008_002869
+2008_002870
+2008_002872
+2008_002873
+2008_002875
+2008_002876
+2008_002879
+2008_002880
+2008_002882
+2008_002883
+2008_002885
+2008_002887
+2008_002890
+2008_002891
+2008_002892
+2008_002894
+2008_002897
+2008_002899
+2008_002900
+2008_002903
+2008_002904
+2008_002906
+2008_002908
+2008_002909
+2008_002910
+2008_002913
+2008_002916
+2008_002917
+2008_002920
+2008_002922
+2008_002926
+2008_002929
+2008_002930
+2008_002931
+2008_002932
+2008_002936
+2008_002942
+2008_002943
+2008_002946
+2008_002947
+2008_002948
+2008_002951
+2008_002954
+2008_002955
+2008_002956
+2008_002957
+2008_002958
+2008_002960
+2008_002961
+2008_002965
+2008_002966
+2008_002968
+2008_002970
+2008_002971
+2008_002972
+2008_002973
+2008_002977
+2008_002983
+2008_002984
+2008_002985
+2008_002988
+2008_002992
+2008_002993
+2008_002997
+2008_002999
+2008_003001
+2008_003003
+2008_003005
+2008_003008
+2008_003013
+2008_003015
+2008_003017
+2008_003018
+2008_003020
+2008_003021
+2008_003022
+2008_003023
+2008_003025
+2008_003026
+2008_003030
+2008_003033
+2008_003034
+2008_003037
+2008_003039
+2008_003041
+2008_003043
+2008_003045
+2008_003048
+2008_003049
+2008_003051
+2008_003052
+2008_003053
+2008_003055
+2008_003056
+2008_003057
+2008_003059
+2008_003060
+2008_003061
+2008_003062
+2008_003063
+2008_003065
+2008_003067
+2008_003068
+2008_003072
+2008_003073
+2008_003075
+2008_003076
+2008_003079
+2008_003081
+2008_003082
+2008_003083
+2008_003087
+2008_003088
+2008_003089
+2008_003090
+2008_003093
+2008_003094
+2008_003095
+2008_003099
+2008_003100
+2008_003101
+2008_003104
+2008_003105
+2008_003106
+2008_003107
+2008_003108
+2008_003110
+2008_003112
+2008_003114
+2008_003120
+2008_003122
+2008_003127
+2008_003128
+2008_003132
+2008_003133
+2008_003134
+2008_003135
+2008_003136
+2008_003140
+2008_003141
+2008_003143
+2008_003144
+2008_003146
+2008_003147
+2008_003151
+2008_003152
+2008_003154
+2008_003155
+2008_003157
+2008_003160
+2008_003161
+2008_003167
+2008_003168
+2008_003170
+2008_003178
+2008_003180
+2008_003181
+2008_003182
+2008_003186
+2008_003187
+2008_003189
+2008_003191
+2008_003193
+2008_003196
+2008_003200
+2008_003202
+2008_003203
+2008_003205
+2008_003208
+2008_003209
+2008_003210
+2008_003211
+2008_003213
+2008_003220
+2008_003222
+2008_003224
+2008_003225
+2008_003228
+2008_003231
+2008_003232
+2008_003238
+2008_003239
+2008_003242
+2008_003244
+2008_003245
+2008_003248
+2008_003249
+2008_003251
+2008_003252
+2008_003255
+2008_003256
+2008_003261
+2008_003263
+2008_003264
+2008_003265
+2008_003266
+2008_003269
+2008_003270
+2008_003271
+2008_003272
+2008_003275
+2008_003276
+2008_003277
+2008_003278
+2008_003280
+2008_003283
+2008_003286
+2008_003287
+2008_003288
+2008_003289
+2008_003290
+2008_003291
+2008_003295
+2008_003297
+2008_003300
+2008_003302
+2008_003303
+2008_003304
+2008_003305
+2008_003311
+2008_003313
+2008_003316
+2008_003318
+2008_003320
+2008_003321
+2008_003323
+2008_003326
+2008_003329
+2008_003330
+2008_003331
+2008_003333
+2008_003334
+2008_003335
+2008_003336
+2008_003338
+2008_003342
+2008_003343
+2008_003344
+2008_003347
+2008_003348
+2008_003350
+2008_003351
+2008_003359
+2008_003360
+2008_003361
+2008_003362
+2008_003369
+2008_003373
+2008_003374
+2008_003378
+2008_003379
+2008_003380
+2008_003381
+2008_003382
+2008_003384
+2008_003386
+2008_003393
+2008_003394
+2008_003395
+2008_003402
+2008_003405
+2008_003406
+2008_003407
+2008_003409
+2008_003414
+2008_003415
+2008_003417
+2008_003418
+2008_003420
+2008_003423
+2008_003424
+2008_003426
+2008_003429
+2008_003430
+2008_003432
+2008_003433
+2008_003434
+2008_003435
+2008_003437
+2008_003439
+2008_003442
+2008_003443
+2008_003447
+2008_003448
+2008_003449
+2008_003451
+2008_003452
+2008_003453
+2008_003458
+2008_003461
+2008_003462
+2008_003463
+2008_003464
+2008_003466
+2008_003467
+2008_003469
+2008_003472
+2008_003475
+2008_003476
+2008_003477
+2008_003478
+2008_003479
+2008_003480
+2008_003482
+2008_003483
+2008_003484
+2008_003485
+2008_003488
+2008_003489
+2008_003492
+2008_003493
+2008_003496
+2008_003497
+2008_003498
+2008_003499
+2008_003500
+2008_003501
+2008_003504
+2008_003507
+2008_003510
+2008_003511
+2008_003514
+2008_003515
+2008_003519
+2008_003520
+2008_003521
+2008_003522
+2008_003523
+2008_003524
+2008_003526
+2008_003531
+2008_003533
+2008_003534
+2008_003542
+2008_003544
+2008_003545
+2008_003546
+2008_003547
+2008_003552
+2008_003557
+2008_003559
+2008_003560
+2008_003562
+2008_003565
+2008_003571
+2008_003572
+2008_003575
+2008_003576
+2008_003577
+2008_003578
+2008_003579
+2008_003580
+2008_003582
+2008_003585
+2008_003587
+2008_003589
+2008_003590
+2008_003591
+2008_003592
+2008_003593
+2008_003596
+2008_003598
+2008_003604
+2008_003607
+2008_003608
+2008_003609
+2008_003610
+2008_003611
+2008_003613
+2008_003617
+2008_003618
+2008_003619
+2008_003621
+2008_003622
+2008_003624
+2008_003626
+2008_003629
+2008_003635
+2008_003636
+2008_003637
+2008_003638
+2008_003645
+2008_003647
+2008_003650
+2008_003652
+2008_003653
+2008_003655
+2008_003658
+2008_003659
+2008_003662
+2008_003665
+2008_003667
+2008_003671
+2008_003672
+2008_003673
+2008_003674
+2008_003675
+2008_003676
+2008_003677
+2008_003680
+2008_003681
+2008_003682
+2008_003683
+2008_003684
+2008_003685
+2008_003688
+2008_003689
+2008_003691
+2008_003694
+2008_003697
+2008_003701
+2008_003703
+2008_003704
+2008_003706
+2008_003707
+2008_003709
+2008_003712
+2008_003713
+2008_003718
+2008_003719
+2008_003720
+2008_003721
+2008_003722
+2008_003726
+2008_003729
+2008_003732
+2008_003733
+2008_003737
+2008_003743
+2008_003744
+2008_003745
+2008_003746
+2008_003748
+2008_003749
+2008_003753
+2008_003754
+2008_003755
+2008_003756
+2008_003761
+2008_003762
+2008_003763
+2008_003764
+2008_003766
+2008_003767
+2008_003768
+2008_003769
+2008_003772
+2008_003773
+2008_003774
+2008_003775
+2008_003776
+2008_003777
+2008_003779
+2008_003780
+2008_003781
+2008_003782
+2008_003788
+2008_003789
+2008_003791
+2008_003793
+2008_003794
+2008_003796
+2008_003799
+2008_003800
+2008_003801
+2008_003802
+2008_003805
+2008_003811
+2008_003812
+2008_003813
+2008_003814
+2008_003815
+2008_003819
+2008_003820
+2008_003821
+2008_003825
+2008_003826
+2008_003827
+2008_003829
+2008_003830
+2008_003831
+2008_003835
+2008_003838
+2008_003840
+2008_003841
+2008_003842
+2008_003843
+2008_003844
+2008_003846
+2008_003847
+2008_003849
+2008_003852
+2008_003854
+2008_003856
+2008_003858
+2008_003860
+2008_003864
+2008_003866
+2008_003868
+2008_003870
+2008_003871
+2008_003873
+2008_003874
+2008_003876
+2008_003881
+2008_003882
+2008_003883
+2008_003884
+2008_003885
+2008_003886
+2008_003888
+2008_003891
+2008_003892
+2008_003894
+2008_003904
+2008_003905
+2008_003908
+2008_003913
+2008_003914
+2008_003915
+2008_003916
+2008_003920
+2008_003921
+2008_003922
+2008_003924
+2008_003925
+2008_003926
+2008_003929
+2008_003932
+2008_003933
+2008_003939
+2008_003940
+2008_003941
+2008_003942
+2008_003943
+2008_003944
+2008_003945
+2008_003947
+2008_003948
+2008_003951
+2008_003956
+2008_003958
+2008_003962
+2008_003965
+2008_003966
+2008_003967
+2008_003969
+2008_003970
+2008_003971
+2008_003974
+2008_003975
+2008_003976
+2008_003978
+2008_003983
+2008_003984
+2008_003985
+2008_003986
+2008_003988
+2008_003989
+2008_003992
+2008_003995
+2008_003996
+2008_003997
+2008_003998
+2008_004000
+2008_004002
+2008_004003
+2008_004004
+2008_004006
+2008_004007
+2008_004008
+2008_004014
+2008_004015
+2008_004016
+2008_004017
+2008_004018
+2008_004020
+2008_004021
+2008_004022
+2008_004024
+2008_004026
+2008_004027
+2008_004030
+2008_004036
+2008_004037
+2008_004040
+2008_004042
+2008_004044
+2008_004045
+2008_004046
+2008_004048
+2008_004053
+2008_004054
+2008_004055
+2008_004056
+2008_004058
+2008_004064
+2008_004066
+2008_004069
+2008_004071
+2008_004074
+2008_004075
+2008_004076
+2008_004077
+2008_004080
+2008_004081
+2008_004084
+2008_004087
+2008_004088
+2008_004090
+2008_004092
+2008_004093
+2008_004097
+2008_004100
+2008_004101
+2008_004102
+2008_004103
+2008_004105
+2008_004106
+2008_004110
+2008_004112
+2008_004113
+2008_004119
+2008_004120
+2008_004121
+2008_004122
+2008_004123
+2008_004124
+2008_004125
+2008_004126
+2008_004127
+2008_004130
+2008_004134
+2008_004135
+2008_004137
+2008_004138
+2008_004140
+2008_004142
+2008_004145
+2008_004147
+2008_004148
+2008_004155
+2008_004161
+2008_004163
+2008_004165
+2008_004166
+2008_004171
+2008_004174
+2008_004175
+2008_004176
+2008_004178
+2008_004182
+2008_004188
+2008_004189
+2008_004190
+2008_004195
+2008_004196
+2008_004198
+2008_004201
+2008_004203
+2008_004205
+2008_004208
+2008_004212
+2008_004213
+2008_004214
+2008_004216
+2008_004217
+2008_004218
+2008_004221
+2008_004224
+2008_004230
+2008_004231
+2008_004232
+2008_004234
+2008_004235
+2008_004239
+2008_004242
+2008_004243
+2008_004245
+2008_004246
+2008_004247
+2008_004251
+2008_004257
+2008_004258
+2008_004259
+2008_004263
+2008_004265
+2008_004269
+2008_004270
+2008_004271
+2008_004273
+2008_004274
+2008_004276
+2008_004278
+2008_004279
+2008_004280
+2008_004284
+2008_004287
+2008_004288
+2008_004289
+2008_004290
+2008_004291
+2008_004292
+2008_004293
+2008_004296
+2008_004297
+2008_004301
+2008_004303
+2008_004306
+2008_004307
+2008_004308
+2008_004312
+2008_004313
+2008_004314
+2008_004317
+2008_004318
+2008_004319
+2008_004321
+2008_004324
+2008_004325
+2008_004326
+2008_004327
+2008_004328
+2008_004330
+2008_004331
+2008_004333
+2008_004339
+2008_004342
+2008_004344
+2008_004345
+2008_004347
+2008_004348
+2008_004353
+2008_004354
+2008_004357
+2008_004358
+2008_004361
+2008_004362
+2008_004363
+2008_004365
+2008_004367
+2008_004371
+2008_004372
+2008_004374
+2008_004376
+2008_004378
+2008_004380
+2008_004384
+2008_004385
+2008_004387
+2008_004389
+2008_004391
+2008_004394
+2008_004396
+2008_004398
+2008_004399
+2008_004402
+2008_004403
+2008_004406
+2008_004408
+2008_004410
+2008_004411
+2008_004412
+2008_004414
+2008_004416
+2008_004417
+2008_004418
+2008_004419
+2008_004422
+2008_004425
+2008_004426
+2008_004427
+2008_004428
+2008_004430
+2008_004431
+2008_004433
+2008_004435
+2008_004436
+2008_004438
+2008_004439
+2008_004441
+2008_004443
+2008_004445
+2008_004450
+2008_004452
+2008_004453
+2008_004455
+2008_004457
+2008_004458
+2008_004459
+2008_004460
+2008_004462
+2008_004464
+2008_004469
+2008_004470
+2008_004471
+2008_004476
+2008_004477
+2008_004478
+2008_004479
+2008_004480
+2008_004482
+2008_004487
+2008_004488
+2008_004490
+2008_004492
+2008_004493
+2008_004497
+2008_004498
+2008_004499
+2008_004501
+2008_004502
+2008_004504
+2008_004505
+2008_004506
+2008_004510
+2008_004512
+2008_004513
+2008_004515
+2008_004518
+2008_004519
+2008_004520
+2008_004522
+2008_004525
+2008_004526
+2008_004528
+2008_004532
+2008_004533
+2008_004534
+2008_004538
+2008_004539
+2008_004540
+2008_004541
+2008_004544
+2008_004545
+2008_004546
+2008_004547
+2008_004549
+2008_004550
+2008_004551
+2008_004552
+2008_004553
+2008_004554
+2008_004559
+2008_004564
+2008_004567
+2008_004568
+2008_004570
+2008_004574
+2008_004575
+2008_004579
+2008_004581
+2008_004583
+2008_004584
+2008_004585
+2008_004588
+2008_004589
+2008_004590
+2008_004592
+2008_004593
+2008_004599
+2008_004602
+2008_004603
+2008_004605
+2008_004606
+2008_004607
+2008_004610
+2008_004611
+2008_004612
+2008_004613
+2008_004614
+2008_004615
+2008_004616
+2008_004617
+2008_004619
+2008_004620
+2008_004621
+2008_004624
+2008_004629
+2008_004630
+2008_004631
+2008_004632
+2008_004633
+2008_004634
+2008_004635
+2008_004636
+2008_004640
+2008_004646
+2008_004647
+2008_004648
+2008_004649
+2008_004653
+2008_004654
+2008_004656
+2008_004659
+2008_004661
+2008_004662
+2008_004663
+2008_004665
+2008_004666
+2008_004667
+2008_004668
+2008_004670
+2008_004671
+2008_004672
+2008_004677
+2008_004678
+2008_004679
+2008_004684
+2008_004687
+2008_004688
+2008_004689
+2008_004690
+2008_004692
+2008_004695
+2008_004696
+2008_004697
+2008_004701
+2008_004702
+2008_004703
+2008_004704
+2008_004705
+2008_004706
+2008_004707
+2008_004711
+2008_004713
+2008_004716
+2008_004718
+2008_004719
+2008_004720
+2008_004722
+2008_004725
+2008_004726
+2008_004729
+2008_004730
+2008_004732
+2008_004736
+2008_004739
+2008_004740
+2008_004742
+2008_004745
+2008_004749
+2008_004750
+2008_004752
+2008_004754
+2008_004756
+2008_004758
+2008_004760
+2008_004763
+2008_004764
+2008_004766
+2008_004767
+2008_004768
+2008_004770
+2008_004771
+2008_004774
+2008_004776
+2008_004777
+2008_004778
+2008_004781
+2008_004783
+2008_004784
+2008_004786
+2008_004794
+2008_004795
+2008_004797
+2008_004802
+2008_004804
+2008_004805
+2008_004807
+2008_004808
+2008_004812
+2008_004814
+2008_004819
+2008_004821
+2008_004822
+2008_004825
+2008_004827
+2008_004832
+2008_004833
+2008_004834
+2008_004837
+2008_004838
+2008_004841
+2008_004844
+2008_004845
+2008_004847
+2008_004849
+2008_004850
+2008_004851
+2008_004852
+2008_004854
+2008_004856
+2008_004858
+2008_004862
+2008_004866
+2008_004868
+2008_004869
+2008_004872
+2008_004873
+2008_004874
+2008_004875
+2008_004876
+2008_004881
+2008_004885
+2008_004887
+2008_004892
+2008_004893
+2008_004894
+2008_004896
+2008_004898
+2008_004899
+2008_004900
+2008_004903
+2008_004904
+2008_004907
+2008_004908
+2008_004910
+2008_004911
+2008_004914
+2008_004917
+2008_004920
+2008_004921
+2008_004923
+2008_004926
+2008_004930
+2008_004931
+2008_004933
+2008_004934
+2008_004935
+2008_004937
+2008_004938
+2008_004940
+2008_004942
+2008_004945
+2008_004946
+2008_004948
+2008_004950
+2008_004955
+2008_004961
+2008_004964
+2008_004966
+2008_004967
+2008_004968
+2008_004969
+2008_004970
+2008_004973
+2008_004974
+2008_004975
+2008_004976
+2008_004977
+2008_004979
+2008_004981
+2008_004982
+2008_004983
+2008_004984
+2008_004985
+2008_004986
+2008_004990
+2008_004991
+2008_004995
+2008_004998
+2008_005000
+2008_005001
+2008_005003
+2008_005006
+2008_005008
+2008_005010
+2008_005013
+2008_005015
+2008_005016
+2008_005023
+2008_005032
+2008_005033
+2008_005035
+2008_005036
+2008_005037
+2008_005040
+2008_005042
+2008_005043
+2008_005045
+2008_005046
+2008_005049
+2008_005051
+2008_005054
+2008_005055
+2008_005057
+2008_005061
+2008_005063
+2008_005064
+2008_005065
+2008_005066
+2008_005068
+2008_005070
+2008_005071
+2008_005072
+2008_005074
+2008_005078
+2008_005080
+2008_005081
+2008_005082
+2008_005084
+2008_005085
+2008_005088
+2008_005089
+2008_005090
+2008_005092
+2008_005094
+2008_005096
+2008_005097
+2008_005098
+2008_005101
+2008_005105
+2008_005107
+2008_005108
+2008_005109
+2008_005110
+2008_005111
+2008_005114
+2008_005115
+2008_005117
+2008_005123
+2008_005127
+2008_005132
+2008_005133
+2008_005134
+2008_005136
+2008_005137
+2008_005139
+2008_005140
+2008_005146
+2008_005147
+2008_005150
+2008_005151
+2008_005156
+2008_005158
+2008_005159
+2008_005160
+2008_005166
+2008_005167
+2008_005168
+2008_005171
+2008_005172
+2008_005174
+2008_005175
+2008_005178
+2008_005181
+2008_005182
+2008_005183
+2008_005185
+2008_005186
+2008_005190
+2008_005191
+2008_005193
+2008_005194
+2008_005196
+2008_005197
+2008_005201
+2008_005204
+2008_005205
+2008_005208
+2008_005209
+2008_005213
+2008_005214
+2008_005215
+2008_005216
+2008_005217
+2008_005218
+2008_005220
+2008_005221
+2008_005231
+2008_005233
+2008_005234
+2008_005235
+2008_005236
+2008_005240
+2008_005242
+2008_005243
+2008_005244
+2008_005245
+2008_005247
+2008_005248
+2008_005250
+2008_005251
+2008_005252
+2008_005253
+2008_005254
+2008_005255
+2008_005257
+2008_005260
+2008_005261
+2008_005266
+2008_005269
+2008_005270
+2008_005271
+2008_005272
+2008_005276
+2008_005277
+2008_005279
+2008_005281
+2008_005282
+2008_005283
+2008_005288
+2008_005294
+2008_005295
+2008_005296
+2008_005297
+2008_005300
+2008_005303
+2008_005304
+2008_005309
+2008_005310
+2008_005313
+2008_005315
+2008_005316
+2008_005319
+2008_005321
+2008_005323
+2008_005324
+2008_005325
+2008_005327
+2008_005329
+2008_005331
+2008_005333
+2008_005335
+2008_005336
+2008_005337
+2008_005338
+2008_005342
+2008_005345
+2008_005346
+2008_005347
+2008_005348
+2008_005349
+2008_005350
+2008_005354
+2008_005356
+2008_005357
+2008_005359
+2008_005360
+2008_005361
+2008_005362
+2008_005363
+2008_005365
+2008_005367
+2008_005369
+2008_005373
+2008_005374
+2008_005375
+2008_005376
+2008_005378
+2008_005379
+2008_005380
+2008_005382
+2008_005386
+2008_005389
+2008_005393
+2008_005395
+2008_005396
+2008_005398
+2008_005399
+2008_005400
+2008_005404
+2008_005405
+2008_005406
+2008_005408
+2008_005412
+2008_005414
+2008_005415
+2008_005417
+2008_005421
+2008_005422
+2008_005423
+2008_005427
+2008_005429
+2008_005431
+2008_005433
+2008_005436
+2008_005439
+2008_005443
+2008_005444
+2008_005445
+2008_005446
+2008_005447
+2008_005449
+2008_005451
+2008_005455
+2008_005456
+2008_005460
+2008_005463
+2008_005465
+2008_005467
+2008_005469
+2008_005472
+2008_005473
+2008_005477
+2008_005480
+2008_005484
+2008_005485
+2008_005490
+2008_005491
+2008_005494
+2008_005496
+2008_005498
+2008_005500
+2008_005501
+2008_005502
+2008_005504
+2008_005505
+2008_005507
+2008_005510
+2008_005511
+2008_005512
+2008_005514
+2008_005517
+2008_005519
+2008_005521
+2008_005522
+2008_005523
+2008_005525
+2008_005526
+2008_005527
+2008_005530
+2008_005531
+2008_005534
+2008_005536
+2008_005538
+2008_005541
+2008_005544
+2008_005548
+2008_005549
+2008_005550
+2008_005552
+2008_005553
+2008_005558
+2008_005560
+2008_005561
+2008_005563
+2008_005564
+2008_005566
+2008_005567
+2008_005569
+2008_005570
+2008_005572
+2008_005573
+2008_005574
+2008_005582
+2008_005584
+2008_005588
+2008_005589
+2008_005591
+2008_005593
+2008_005599
+2008_005600
+2008_005601
+2008_005603
+2008_005608
+2008_005609
+2008_005610
+2008_005611
+2008_005612
+2008_005614
+2008_005616
+2008_005618
+2008_005623
+2008_005625
+2008_005626
+2008_005627
+2008_005628
+2008_005631
+2008_005633
+2008_005634
+2008_005635
+2008_005636
+2008_005637
+2008_005638
+2008_005639
+2008_005641
+2008_005642
+2008_005643
+2008_005646
+2008_005649
+2008_005650
+2008_005652
+2008_005653
+2008_005656
+2008_005657
+2008_005660
+2008_005663
+2008_005664
+2008_005668
+2008_005673
+2008_005675
+2008_005676
+2008_005677
+2008_005678
+2008_005679
+2008_005680
+2008_005681
+2008_005682
+2008_005683
+2008_005685
+2008_005686
+2008_005687
+2008_005691
+2008_005695
+2008_005698
+2008_005699
+2008_005701
+2008_005702
+2008_005703
+2008_005705
+2008_005706
+2008_005707
+2008_005713
+2008_005714
+2008_005716
+2008_005719
+2008_005720
+2008_005721
+2008_005724
+2008_005726
+2008_005727
+2008_005728
+2008_005732
+2008_005734
+2008_005735
+2008_005736
+2008_005737
+2008_005738
+2008_005739
+2008_005742
+2008_005747
+2008_005748
+2008_005750
+2008_005752
+2008_005757
+2008_005758
+2008_005761
+2008_005763
+2008_005764
+2008_005767
+2008_005768
+2008_005770
+2008_005774
+2008_005777
+2008_005779
+2008_005780
+2008_005788
+2008_005790
+2008_005791
+2008_005792
+2008_005794
+2008_005796
+2008_005798
+2008_005800
+2008_005801
+2008_005803
+2008_005805
+2008_005808
+2008_005810
+2008_005812
+2008_005816
+2008_005817
+2008_005818
+2008_005821
+2008_005822
+2008_005823
+2008_005825
+2008_005831
+2008_005832
+2008_005834
+2008_005838
+2008_005839
+2008_005843
+2008_005845
+2008_005846
+2008_005847
+2008_005848
+2008_005850
+2008_005853
+2008_005855
+2008_005856
+2008_005857
+2008_005860
+2008_005863
+2008_005865
+2008_005867
+2008_005869
+2008_005871
+2008_005873
+2008_005874
+2008_005875
+2008_005877
+2008_005878
+2008_005881
+2008_005882
+2008_005883
+2008_005884
+2008_005889
+2008_005890
+2008_005891
+2008_005893
+2008_005895
+2008_005897
+2008_005898
+2008_005902
+2008_005903
+2008_005904
+2008_005905
+2008_005907
+2008_005914
+2008_005915
+2008_005916
+2008_005918
+2008_005921
+2008_005923
+2008_005924
+2008_005926
+2008_005928
+2008_005929
+2008_005933
+2008_005934
+2008_005935
+2008_005936
+2008_005937
+2008_005938
+2008_005939
+2008_005943
+2008_005945
+2008_005954
+2008_005956
+2008_005957
+2008_005959
+2008_005960
+2008_005962
+2008_005964
+2008_005967
+2008_005968
+2008_005970
+2008_005972
+2008_005975
+2008_005976
+2008_005977
+2008_005978
+2008_005979
+2008_005980
+2008_005982
+2008_005984
+2008_005987
+2008_005989
+2008_005991
+2008_005997
+2008_006000
+2008_006002
+2008_006004
+2008_006007
+2008_006008
+2008_006010
+2008_006014
+2008_006017
+2008_006020
+2008_006021
+2008_006024
+2008_006027
+2008_006028
+2008_006031
+2008_006032
+2008_006034
+2008_006036
+2008_006037
+2008_006038
+2008_006039
+2008_006041
+2008_006042
+2008_006045
+2008_006046
+2008_006047
+2008_006049
+2008_006050
+2008_006052
+2008_006055
+2008_006058
+2008_006059
+2008_006062
+2008_006063
+2008_006064
+2008_006065
+2008_006067
+2008_006068
+2008_006070
+2008_006071
+2008_006072
+2008_006074
+2008_006076
+2008_006078
+2008_006081
+2008_006082
+2008_006085
+2008_006087
+2008_006088
+2008_006090
+2008_006092
+2008_006094
+2008_006096
+2008_006099
+2008_006100
+2008_006102
+2008_006104
+2008_006108
+2008_006109
+2008_006111
+2008_006112
+2008_006113
+2008_006117
+2008_006119
+2008_006120
+2008_006121
+2008_006124
+2008_006128
+2008_006129
+2008_006130
+2008_006133
+2008_006135
+2008_006136
+2008_006140
+2008_006143
+2008_006144
+2008_006145
+2008_006147
+2008_006148
+2008_006151
+2008_006152
+2008_006154
+2008_006158
+2008_006159
+2008_006163
+2008_006164
+2008_006166
+2008_006169
+2008_006170
+2008_006175
+2008_006178
+2008_006179
+2008_006181
+2008_006182
+2008_006185
+2008_006186
+2008_006188
+2008_006190
+2008_006192
+2008_006194
+2008_006195
+2008_006200
+2008_006203
+2008_006205
+2008_006207
+2008_006210
+2008_006211
+2008_006213
+2008_006215
+2008_006216
+2008_006218
+2008_006219
+2008_006220
+2008_006221
+2008_006222
+2008_006224
+2008_006225
+2008_006227
+2008_006229
+2008_006232
+2008_006233
+2008_006234
+2008_006235
+2008_006239
+2008_006240
+2008_006242
+2008_006244
+2008_006249
+2008_006250
+2008_006253
+2008_006254
+2008_006256
+2008_006257
+2008_006258
+2008_006262
+2008_006265
+2008_006267
+2008_006269
+2008_006271
+2008_006272
+2008_006273
+2008_006275
+2008_006276
+2008_006280
+2008_006281
+2008_006282
+2008_006285
+2008_006288
+2008_006289
+2008_006290
+2008_006294
+2008_006295
+2008_006298
+2008_006300
+2008_006303
+2008_006307
+2008_006310
+2008_006311
+2008_006315
+2008_006316
+2008_006317
+2008_006320
+2008_006323
+2008_006325
+2008_006327
+2008_006329
+2008_006330
+2008_006331
+2008_006335
+2008_006336
+2008_006337
+2008_006339
+2008_006341
+2008_006345
+2008_006347
+2008_006349
+2008_006350
+2008_006351
+2008_006353
+2008_006355
+2008_006356
+2008_006359
+2008_006361
+2008_006362
+2008_006364
+2008_006365
+2008_006366
+2008_006368
+2008_006369
+2008_006370
+2008_006373
+2008_006376
+2008_006377
+2008_006382
+2008_006384
+2008_006386
+2008_006387
+2008_006389
+2008_006390
+2008_006392
+2008_006394
+2008_006397
+2008_006400
+2008_006401
+2008_006403
+2008_006404
+2008_006407
+2008_006408
+2008_006409
+2008_006410
+2008_006416
+2008_006417
+2008_006419
+2008_006421
+2008_006424
+2008_006425
+2008_006427
+2008_006429
+2008_006430
+2008_006432
+2008_006433
+2008_006434
+2008_006436
+2008_006438
+2008_006441
+2008_006447
+2008_006448
+2008_006449
+2008_006452
+2008_006458
+2008_006461
+2008_006462
+2008_006463
+2008_006467
+2008_006470
+2008_006474
+2008_006475
+2008_006477
+2008_006480
+2008_006481
+2008_006482
+2008_006483
+2008_006487
+2008_006488
+2008_006489
+2008_006490
+2008_006491
+2008_006496
+2008_006497
+2008_006500
+2008_006502
+2008_006503
+2008_006506
+2008_006509
+2008_006511
+2008_006512
+2008_006517
+2008_006519
+2008_006520
+2008_006522
+2008_006523
+2008_006524
+2008_006526
+2008_006528
+2008_006530
+2008_006534
+2008_006538
+2008_006540
+2008_006543
+2008_006546
+2008_006547
+2008_006548
+2008_006549
+2008_006553
+2008_006554
+2008_006558
+2008_006561
+2008_006562
+2008_006564
+2008_006566
+2008_006567
+2008_006568
+2008_006570
+2008_006576
+2008_006578
+2008_006579
+2008_006585
+2008_006586
+2008_006587
+2008_006588
+2008_006591
+2008_006598
+2008_006599
+2008_006600
+2008_006602
+2008_006604
+2008_006605
+2008_006606
+2008_006609
+2008_006610
+2008_006611
+2008_006613
+2008_006614
+2008_006616
+2008_006617
+2008_006619
+2008_006621
+2008_006623
+2008_006624
+2008_006625
+2008_006626
+2008_006629
+2008_006631
+2008_006634
+2008_006635
+2008_006637
+2008_006638
+2008_006641
+2008_006642
+2008_006645
+2008_006646
+2008_006649
+2008_006650
+2008_006654
+2008_006655
+2008_006656
+2008_006657
+2008_006660
+2008_006662
+2008_006663
+2008_006665
+2008_006667
+2008_006668
+2008_006671
+2008_006677
+2008_006682
+2008_006684
+2008_006686
+2008_006690
+2008_006691
+2008_006692
+2008_006694
+2008_006696
+2008_006700
+2008_006701
+2008_006703
+2008_006705
+2008_006708
+2008_006710
+2008_006712
+2008_006714
+2008_006715
+2008_006716
+2008_006717
+2008_006718
+2008_006719
+2008_006720
+2008_006722
+2008_006724
+2008_006728
+2008_006730
+2008_006731
+2008_006732
+2008_006733
+2008_006737
+2008_006743
+2008_006746
+2008_006747
+2008_006748
+2008_006750
+2008_006751
+2008_006752
+2008_006753
+2008_006758
+2008_006761
+2008_006762
+2008_006764
+2008_006765
+2008_006767
+2008_006773
+2008_006774
+2008_006776
+2008_006777
+2008_006778
+2008_006779
+2008_006781
+2008_006784
+2008_006785
+2008_006792
+2008_006793
+2008_006796
+2008_006797
+2008_006798
+2008_006800
+2008_006802
+2008_006807
+2008_006808
+2008_006810
+2008_006811
+2008_006813
+2008_006815
+2008_006816
+2008_006817
+2008_006818
+2008_006819
+2008_006820
+2008_006824
+2008_006825
+2008_006827
+2008_006828
+2008_006831
+2008_006832
+2008_006833
+2008_006834
+2008_006835
+2008_006837
+2008_006839
+2008_006841
+2008_006843
+2008_006844
+2008_006847
+2008_006849
+2008_006855
+2008_006857
+2008_006863
+2008_006864
+2008_006865
+2008_006868
+2008_006870
+2008_006872
+2008_006873
+2008_006874
+2008_006877
+2008_006879
+2008_006880
+2008_006881
+2008_006882
+2008_006885
+2008_006887
+2008_006889
+2008_006890
+2008_006892
+2008_006896
+2008_006898
+2008_006900
+2008_006902
+2008_006903
+2008_006904
+2008_006907
+2008_006908
+2008_006909
+2008_006910
+2008_006912
+2008_006919
+2008_006920
+2008_006921
+2008_006923
+2008_006924
+2008_006925
+2008_006926
+2008_006933
+2008_006936
+2008_006939
+2008_006941
+2008_006944
+2008_006946
+2008_006948
+2008_006949
+2008_006950
+2008_006951
+2008_006952
+2008_006953
+2008_006954
+2008_006956
+2008_006959
+2008_006960
+2008_006961
+2008_006962
+2008_006965
+2008_006967
+2008_006968
+2008_006969
+2008_006973
+2008_006979
+2008_006980
+2008_006981
+2008_006986
+2008_006987
+2008_006989
+2008_006991
+2008_006992
+2008_006997
+2008_006998
+2008_006999
+2008_007003
+2008_007004
+2008_007006
+2008_007009
+2008_007010
+2008_007011
+2008_007012
+2008_007014
+2008_007019
+2008_007021
+2008_007022
+2008_007025
+2008_007026
+2008_007028
+2008_007030
+2008_007031
+2008_007032
+2008_007034
+2008_007038
+2008_007039
+2008_007042
+2008_007043
+2008_007045
+2008_007048
+2008_007050
+2008_007054
+2008_007056
+2008_007057
+2008_007058
+2008_007059
+2008_007060
+2008_007061
+2008_007064
+2008_007067
+2008_007069
+2008_007070
+2008_007073
+2008_007075
+2008_007076
+2008_007081
+2008_007082
+2008_007084
+2008_007085
+2008_007086
+2008_007090
+2008_007091
+2008_007095
+2008_007096
+2008_007097
+2008_007098
+2008_007101
+2008_007103
+2008_007105
+2008_007106
+2008_007108
+2008_007112
+2008_007114
+2008_007115
+2008_007118
+2008_007119
+2008_007120
+2008_007123
+2008_007124
+2008_007129
+2008_007130
+2008_007131
+2008_007133
+2008_007134
+2008_007138
+2008_007142
+2008_007143
+2008_007145
+2008_007146
+2008_007147
+2008_007151
+2008_007156
+2008_007161
+2008_007163
+2008_007164
+2008_007165
+2008_007166
+2008_007167
+2008_007168
+2008_007169
+2008_007171
+2008_007176
+2008_007179
+2008_007181
+2008_007182
+2008_007184
+2008_007185
+2008_007187
+2008_007188
+2008_007189
+2008_007190
+2008_007194
+2008_007195
+2008_007196
+2008_007197
+2008_007201
+2008_007205
+2008_007207
+2008_007208
+2008_007211
+2008_007214
+2008_007216
+2008_007217
+2008_007218
+2008_007219
+2008_007221
+2008_007222
+2008_007223
+2008_007225
+2008_007226
+2008_007227
+2008_007229
+2008_007231
+2008_007236
+2008_007237
+2008_007239
+2008_007241
+2008_007242
+2008_007245
+2008_007246
+2008_007247
+2008_007250
+2008_007252
+2008_007254
+2008_007256
+2008_007260
+2008_007261
+2008_007264
+2008_007265
+2008_007266
+2008_007269
+2008_007273
+2008_007274
+2008_007277
+2008_007279
+2008_007280
+2008_007281
+2008_007282
+2008_007285
+2008_007286
+2008_007287
+2008_007289
+2008_007291
+2008_007293
+2008_007295
+2008_007298
+2008_007305
+2008_007307
+2008_007311
+2008_007312
+2008_007313
+2008_007314
+2008_007317
+2008_007319
+2008_007320
+2008_007321
+2008_007323
+2008_007324
+2008_007325
+2008_007327
+2008_007332
+2008_007334
+2008_007335
+2008_007336
+2008_007339
+2008_007343
+2008_007344
+2008_007346
+2008_007348
+2008_007350
+2008_007352
+2008_007353
+2008_007356
+2008_007357
+2008_007358
+2008_007361
+2008_007363
+2008_007364
+2008_007374
+2008_007375
+2008_007378
+2008_007382
+2008_007383
+2008_007384
+2008_007388
+2008_007389
+2008_007390
+2008_007392
+2008_007393
+2008_007394
+2008_007397
+2008_007398
+2008_007402
+2008_007403
+2008_007404
+2008_007409
+2008_007410
+2008_007415
+2008_007417
+2008_007421
+2008_007423
+2008_007424
+2008_007425
+2008_007428
+2008_007430
+2008_007431
+2008_007432
+2008_007433
+2008_007434
+2008_007435
+2008_007438
+2008_007441
+2008_007442
+2008_007443
+2008_007444
+2008_007446
+2008_007448
+2008_007452
+2008_007455
+2008_007456
+2008_007458
+2008_007459
+2008_007461
+2008_007465
+2008_007466
+2008_007469
+2008_007470
+2008_007471
+2008_007472
+2008_007473
+2008_007476
+2008_007477
+2008_007478
+2008_007480
+2008_007485
+2008_007486
+2008_007488
+2008_007491
+2008_007494
+2008_007496
+2008_007497
+2008_007498
+2008_007500
+2008_007501
+2008_007504
+2008_007507
+2008_007509
+2008_007510
+2008_007511
+2008_007513
+2008_007514
+2008_007515
+2008_007519
+2008_007521
+2008_007524
+2008_007525
+2008_007527
+2008_007528
+2008_007529
+2008_007531
+2008_007533
+2008_007534
+2008_007536
+2008_007537
+2008_007538
+2008_007544
+2008_007546
+2008_007548
+2008_007556
+2008_007558
+2008_007559
+2008_007561
+2008_007565
+2008_007567
+2008_007573
+2008_007574
+2008_007576
+2008_007579
+2008_007581
+2008_007583
+2008_007584
+2008_007585
+2008_007586
+2008_007587
+2008_007588
+2008_007589
+2008_007591
+2008_007593
+2008_007594
+2008_007595
+2008_007596
+2008_007597
+2008_007599
+2008_007604
+2008_007608
+2008_007610
+2008_007611
+2008_007612
+2008_007613
+2008_007617
+2008_007618
+2008_007621
+2008_007623
+2008_007625
+2008_007629
+2008_007630
+2008_007632
+2008_007635
+2008_007640
+2008_007641
+2008_007643
+2008_007646
+2008_007648
+2008_007649
+2008_007653
+2008_007656
+2008_007660
+2008_007661
+2008_007662
+2008_007664
+2008_007665
+2008_007666
+2008_007668
+2008_007669
+2008_007673
+2008_007675
+2008_007676
+2008_007677
+2008_007682
+2008_007683
+2008_007685
+2008_007688
+2008_007690
+2008_007691
+2008_007692
+2008_007693
+2008_007694
+2008_007696
+2008_007697
+2008_007698
+2008_007701
+2008_007702
+2008_007704
+2008_007706
+2008_007709
+2008_007710
+2008_007714
+2008_007716
+2008_007717
+2008_007719
+2008_007724
+2008_007726
+2008_007729
+2008_007730
+2008_007733
+2008_007735
+2008_007736
+2008_007737
+2008_007738
+2008_007739
+2008_007741
+2008_007742
+2008_007745
+2008_007746
+2008_007748
+2008_007749
+2008_007750
+2008_007752
+2008_007755
+2008_007757
+2008_007758
+2008_007759
+2008_007760
+2008_007761
+2008_007764
+2008_007766
+2008_007768
+2008_007770
+2008_007777
+2008_007779
+2008_007780
+2008_007781
+2008_007786
+2008_007787
+2008_007788
+2008_007789
+2008_007791
+2008_007793
+2008_007794
+2008_007797
+2008_007798
+2008_007804
+2008_007805
+2008_007806
+2008_007811
+2008_007812
+2008_007814
+2008_007816
+2008_007817
+2008_007819
+2008_007823
+2008_007825
+2008_007827
+2008_007828
+2008_007829
+2008_007831
+2008_007833
+2008_007835
+2008_007836
+2008_007837
+2008_007839
+2008_007840
+2008_007841
+2008_007842
+2008_007843
+2008_007848
+2008_007850
+2008_007852
+2008_007853
+2008_007854
+2008_007855
+2008_007858
+2008_007861
+2008_007864
+2008_007869
+2008_007870
+2008_007871
+2008_007872
+2008_007873
+2008_007875
+2008_007877
+2008_007879
+2008_007882
+2008_007883
+2008_007884
+2008_007887
+2008_007888
+2008_007890
+2008_007891
+2008_007893
+2008_007895
+2008_007897
+2008_007902
+2008_007904
+2008_007907
+2008_007909
+2008_007912
+2008_007913
+2008_007914
+2008_007915
+2008_007916
+2008_007917
+2008_007918
+2008_007922
+2008_007923
+2008_007928
+2008_007931
+2008_007932
+2008_007933
+2008_007935
+2008_007936
+2008_007937
+2008_007938
+2008_007940
+2008_007941
+2008_007942
+2008_007945
+2008_007947
+2008_007948
+2008_007949
+2008_007950
+2008_007953
+2008_007954
+2008_007955
+2008_007962
+2008_007964
+2008_007966
+2008_007969
+2008_007970
+2008_007973
+2008_007975
+2008_007977
+2008_007981
+2008_007985
+2008_007986
+2008_007987
+2008_007988
+2008_007989
+2008_007990
+2008_007993
+2008_007994
+2008_007997
+2008_007998
+2008_007999
+2008_008001
+2008_008002
+2008_008004
+2008_008007
+2008_008011
+2008_008012
+2008_008018
+2008_008020
+2008_008021
+2008_008022
+2008_008024
+2008_008025
+2008_008028
+2008_008029
+2008_008031
+2008_008034
+2008_008037
+2008_008040
+2008_008043
+2008_008044
+2008_008048
+2008_008050
+2008_008052
+2008_008053
+2008_008055
+2008_008057
+2008_008058
+2008_008064
+2008_008066
+2008_008069
+2008_008070
+2008_008072
+2008_008073
+2008_008074
+2008_008075
+2008_008080
+2008_008083
+2008_008084
+2008_008086
+2008_008091
+2008_008092
+2008_008093
+2008_008095
+2008_008096
+2008_008097
+2008_008098
+2008_008103
+2008_008105
+2008_008106
+2008_008109
+2008_008112
+2008_008113
+2008_008115
+2008_008116
+2008_008120
+2008_008121
+2008_008122
+2008_008123
+2008_008125
+2008_008127
+2008_008130
+2008_008131
+2008_008132
+2008_008134
+2008_008141
+2008_008145
+2008_008146
+2008_008147
+2008_008148
+2008_008150
+2008_008152
+2008_008154
+2008_008155
+2008_008162
+2008_008166
+2008_008169
+2008_008170
+2008_008175
+2008_008176
+2008_008177
+2008_008179
+2008_008180
+2008_008184
+2008_008185
+2008_008190
+2008_008191
+2008_008192
+2008_008193
+2008_008194
+2008_008197
+2008_008199
+2008_008200
+2008_008203
+2008_008206
+2008_008208
+2008_008210
+2008_008211
+2008_008212
+2008_008215
+2008_008217
+2008_008218
+2008_008220
+2008_008221
+2008_008223
+2008_008224
+2008_008227
+2008_008229
+2008_008231
+2008_008232
+2008_008233
+2008_008234
+2008_008235
+2008_008237
+2008_008241
+2008_008242
+2008_008246
+2008_008247
+2008_008252
+2008_008254
+2008_008257
+2008_008262
+2008_008263
+2008_008266
+2008_008268
+2008_008269
+2008_008271
+2008_008272
+2008_008274
+2008_008275
+2008_008276
+2008_008278
+2008_008279
+2008_008281
+2008_008284
+2008_008287
+2008_008288
+2008_008292
+2008_008294
+2008_008296
+2008_008297
+2008_008300
+2008_008301
+2008_008302
+2008_008307
+2008_008309
+2008_008310
+2008_008313
+2008_008314
+2008_008315
+2008_008318
+2008_008319
+2008_008320
+2008_008321
+2008_008322
+2008_008323
+2008_008324
+2008_008325
+2008_008330
+2008_008331
+2008_008335
+2008_008336
+2008_008337
+2008_008338
+2008_008341
+2008_008342
+2008_008343
+2008_008344
+2008_008345
+2008_008346
+2008_008347
+2008_008354
+2008_008356
+2008_008357
+2008_008359
+2008_008362
+2008_008363
+2008_008364
+2008_008365
+2008_008366
+2008_008368
+2008_008370
+2008_008373
+2008_008376
+2008_008377
+2008_008379
+2008_008380
+2008_008382
+2008_008384
+2008_008387
+2008_008388
+2008_008391
+2008_008392
+2008_008393
+2008_008395
+2008_008402
+2008_008403
+2008_008404
+2008_008406
+2008_008410
+2008_008411
+2008_008416
+2008_008421
+2008_008423
+2008_008424
+2008_008428
+2008_008429
+2008_008431
+2008_008432
+2008_008433
+2008_008434
+2008_008435
+2008_008437
+2008_008439
+2008_008440
+2008_008443
+2008_008444
+2008_008446
+2008_008447
+2008_008450
+2008_008453
+2008_008455
+2008_008461
+2008_008462
+2008_008464
+2008_008466
+2008_008467
+2008_008469
+2008_008470
+2008_008471
+2008_008474
+2008_008476
+2008_008479
+2008_008480
+2008_008482
+2008_008487
+2008_008488
+2008_008490
+2008_008496
+2008_008497
+2008_008500
+2008_008501
+2008_008506
+2008_008507
+2008_008508
+2008_008511
+2008_008512
+2008_008517
+2008_008519
+2008_008521
+2008_008522
+2008_008523
+2008_008524
+2008_008525
+2008_008526
+2008_008527
+2008_008528
+2008_008530
+2008_008531
+2008_008533
+2008_008536
+2008_008537
+2008_008538
+2008_008541
+2008_008544
+2008_008545
+2008_008546
+2008_008547
+2008_008549
+2008_008550
+2008_008552
+2008_008554
+2008_008560
+2008_008564
+2008_008567
+2008_008570
+2008_008572
+2008_008574
+2008_008578
+2008_008579
+2008_008583
+2008_008585
+2008_008588
+2008_008589
+2008_008590
+2008_008591
+2008_008593
+2008_008595
+2008_008598
+2008_008600
+2008_008601
+2008_008606
+2008_008607
+2008_008608
+2008_008611
+2008_008613
+2008_008615
+2008_008616
+2008_008617
+2008_008618
+2008_008619
+2008_008621
+2008_008622
+2008_008623
+2008_008624
+2008_008627
+2008_008628
+2008_008629
+2008_008632
+2008_008635
+2008_008636
+2008_008637
+2008_008641
+2008_008642
+2008_008649
+2008_008652
+2008_008654
+2008_008658
+2008_008659
+2008_008662
+2008_008665
+2008_008666
+2008_008668
+2008_008671
+2008_008673
+2008_008674
+2008_008675
+2008_008676
+2008_008679
+2008_008681
+2008_008682
+2008_008683
+2008_008684
+2008_008685
+2008_008689
+2008_008690
+2008_008691
+2008_008694
+2008_008695
+2008_008696
+2008_008697
+2008_008700
+2008_008701
+2008_008705
+2008_008706
+2008_008707
+2008_008708
+2008_008711
+2008_008713
+2008_008714
+2008_008717
+2008_008718
+2008_008719
+2008_008724
+2008_008725
+2008_008726
+2008_008732
+2008_008735
+2008_008739
+2008_008744
+2008_008745
+2008_008746
+2008_008748
+2008_008749
+2008_008751
+2008_008753
+2008_008755
+2008_008757
+2008_008758
+2008_008765
+2008_008767
+2008_008770
+2008_008772
+2008_008773
+2009_000001
+2009_000002
+2009_000006
+2009_000009
+2009_000010
+2009_000011
+2009_000012
+2009_000013
+2009_000014
+2009_000015
+2009_000016
+2009_000017
+2009_000021
+2009_000022
+2009_000026
+2009_000027
+2009_000028
+2009_000029
+2009_000030
+2009_000032
+2009_000035
+2009_000037
+2009_000039
+2009_000040
+2009_000041
+2009_000042
+2009_000045
+2009_000051
+2009_000052
+2009_000054
+2009_000055
+2009_000056
+2009_000058
+2009_000059
+2009_000060
+2009_000063
+2009_000066
+2009_000067
+2009_000068
+2009_000072
+2009_000073
+2009_000074
+2009_000078
+2009_000080
+2009_000082
+2009_000084
+2009_000085
+2009_000087
+2009_000088
+2009_000089
+2009_000090
+2009_000091
+2009_000093
+2009_000096
+2009_000097
+2009_000100
+2009_000102
+2009_000103
+2009_000104
+2009_000105
+2009_000109
+2009_000119
+2009_000120
+2009_000121
+2009_000122
+2009_000124
+2009_000128
+2009_000130
+2009_000131
+2009_000132
+2009_000133
+2009_000135
+2009_000136
+2009_000137
+2009_000140
+2009_000141
+2009_000142
+2009_000145
+2009_000146
+2009_000149
+2009_000150
+2009_000151
+2009_000156
+2009_000157
+2009_000158
+2009_000159
+2009_000160
+2009_000161
+2009_000164
+2009_000165
+2009_000168
+2009_000169
+2009_000171
+2009_000176
+2009_000177
+2009_000181
+2009_000182
+2009_000183
+2009_000184
+2009_000188
+2009_000189
+2009_000192
+2009_000195
+2009_000197
+2009_000198
+2009_000199
+2009_000201
+2009_000203
+2009_000205
+2009_000206
+2009_000209
+2009_000212
+2009_000214
+2009_000216
+2009_000217
+2009_000218
+2009_000219
+2009_000223
+2009_000225
+2009_000227
+2009_000229
+2009_000232
+2009_000233
+2009_000237
+2009_000239
+2009_000242
+2009_000244
+2009_000247
+2009_000248
+2009_000249
+2009_000250
+2009_000251
+2009_000253
+2009_000254
+2009_000257
+2009_000260
+2009_000268
+2009_000276
+2009_000277
+2009_000280
+2009_000281
+2009_000282
+2009_000283
+2009_000284
+2009_000285
+2009_000286
+2009_000287
+2009_000288
+2009_000289
+2009_000290
+2009_000291
+2009_000293
+2009_000297
+2009_000298
+2009_000300
+2009_000303
+2009_000304
+2009_000305
+2009_000308
+2009_000309
+2009_000312
+2009_000316
+2009_000317
+2009_000318
+2009_000320
+2009_000321
+2009_000322
+2009_000327
+2009_000328
+2009_000330
+2009_000335
+2009_000336
+2009_000337
+2009_000339
+2009_000340
+2009_000341
+2009_000342
+2009_000343
+2009_000344
+2009_000347
+2009_000350
+2009_000351
+2009_000354
+2009_000356
+2009_000366
+2009_000367
+2009_000370
+2009_000375
+2009_000377
+2009_000378
+2009_000379
+2009_000385
+2009_000387
+2009_000389
+2009_000390
+2009_000391
+2009_000393
+2009_000397
+2009_000398
+2009_000399
+2009_000400
+2009_000402
+2009_000405
+2009_000408
+2009_000409
+2009_000410
+2009_000411
+2009_000412
+2009_000414
+2009_000416
+2009_000417
+2009_000418
+2009_000419
+2009_000420
+2009_000421
+2009_000422
+2009_000426
+2009_000430
+2009_000435
+2009_000438
+2009_000439
+2009_000440
+2009_000443
+2009_000444
+2009_000445
+2009_000446
+2009_000449
+2009_000452
+2009_000453
+2009_000454
+2009_000455
+2009_000456
+2009_000457
+2009_000461
+2009_000463
+2009_000464
+2009_000466
+2009_000469
+2009_000471
+2009_000472
+2009_000474
+2009_000476
+2009_000477
+2009_000483
+2009_000486
+2009_000487
+2009_000488
+2009_000491
+2009_000493
+2009_000494
+2009_000496
+2009_000499
+2009_000500
+2009_000501
+2009_000502
+2009_000503
+2009_000504
+2009_000505
+2009_000511
+2009_000512
+2009_000513
+2009_000515
+2009_000516
+2009_000519
+2009_000522
+2009_000523
+2009_000525
+2009_000526
+2009_000527
+2009_000529
+2009_000532
+2009_000535
+2009_000536
+2009_000539
+2009_000542
+2009_000544
+2009_000545
+2009_000546
+2009_000547
+2009_000549
+2009_000550
+2009_000552
+2009_000553
+2009_000557
+2009_000558
+2009_000559
+2009_000560
+2009_000562
+2009_000563
+2009_000565
+2009_000566
+2009_000567
+2009_000568
+2009_000573
+2009_000574
+2009_000575
+2009_000576
+2009_000577
+2009_000579
+2009_000585
+2009_000586
+2009_000590
+2009_000591
+2009_000592
+2009_000593
+2009_000595
+2009_000597
+2009_000599
+2009_000600
+2009_000602
+2009_000603
+2009_000604
+2009_000606
+2009_000608
+2009_000611
+2009_000614
+2009_000615
+2009_000617
+2009_000619
+2009_000624
+2009_000625
+2009_000626
+2009_000628
+2009_000629
+2009_000631
+2009_000632
+2009_000634
+2009_000635
+2009_000636
+2009_000637
+2009_000638
+2009_000641
+2009_000642
+2009_000647
+2009_000648
+2009_000651
+2009_000653
+2009_000655
+2009_000658
+2009_000661
+2009_000662
+2009_000663
+2009_000664
+2009_000670
+2009_000672
+2009_000674
+2009_000675
+2009_000676
+2009_000677
+2009_000679
+2009_000681
+2009_000683
+2009_000684
+2009_000686
+2009_000689
+2009_000690
+2009_000691
+2009_000692
+2009_000694
+2009_000695
+2009_000696
+2009_000702
+2009_000704
+2009_000705
+2009_000708
+2009_000709
+2009_000712
+2009_000716
+2009_000718
+2009_000719
+2009_000720
+2009_000722
+2009_000723
+2009_000724
+2009_000725
+2009_000726
+2009_000727
+2009_000730
+2009_000731
+2009_000732
+2009_000734
+2009_000737
+2009_000741
+2009_000742
+2009_000744
+2009_000745
+2009_000746
+2009_000748
+2009_000750
+2009_000752
+2009_000755
+2009_000756
+2009_000757
+2009_000758
+2009_000759
+2009_000760
+2009_000762
+2009_000763
+2009_000768
+2009_000770
+2009_000771
+2009_000774
+2009_000777
+2009_000778
+2009_000779
+2009_000782
+2009_000783
+2009_000789
+2009_000790
+2009_000791
+2009_000793
+2009_000794
+2009_000796
+2009_000797
+2009_000801
+2009_000804
+2009_000805
+2009_000811
+2009_000812
+2009_000815
+2009_000816
+2009_000817
+2009_000820
+2009_000821
+2009_000823
+2009_000824
+2009_000825
+2009_000828
+2009_000829
+2009_000830
+2009_000831
+2009_000833
+2009_000834
+2009_000837
+2009_000839
+2009_000840
+2009_000843
+2009_000845
+2009_000846
+2009_000848
+2009_000849
+2009_000851
+2009_000852
+2009_000854
+2009_000856
+2009_000858
+2009_000862
+2009_000865
+2009_000867
+2009_000869
+2009_000871
+2009_000874
+2009_000879
+2009_000882
+2009_000886
+2009_000887
+2009_000889
+2009_000890
+2009_000892
+2009_000894
+2009_000895
+2009_000896
+2009_000897
+2009_000898
+2009_000899
+2009_000901
+2009_000902
+2009_000904
+2009_000906
+2009_000909
+2009_000910
+2009_000915
+2009_000919
+2009_000920
+2009_000923
+2009_000924
+2009_000925
+2009_000926
+2009_000927
+2009_000928
+2009_000930
+2009_000931
+2009_000932
+2009_000934
+2009_000935
+2009_000937
+2009_000938
+2009_000939
+2009_000945
+2009_000948
+2009_000953
+2009_000954
+2009_000955
+2009_000958
+2009_000960
+2009_000961
+2009_000962
+2009_000964
+2009_000966
+2009_000967
+2009_000969
+2009_000970
+2009_000971
+2009_000973
+2009_000974
+2009_000975
+2009_000979
+2009_000980
+2009_000981
+2009_000985
+2009_000987
+2009_000989
+2009_000990
+2009_000991
+2009_000992
+2009_000995
+2009_000996
+2009_000998
+2009_001000
+2009_001002
+2009_001006
+2009_001007
+2009_001008
+2009_001009
+2009_001011
+2009_001012
+2009_001013
+2009_001016
+2009_001019
+2009_001021
+2009_001024
+2009_001026
+2009_001027
+2009_001028
+2009_001030
+2009_001036
+2009_001037
+2009_001038
+2009_001040
+2009_001042
+2009_001044
+2009_001052
+2009_001054
+2009_001055
+2009_001056
+2009_001057
+2009_001059
+2009_001061
+2009_001066
+2009_001068
+2009_001069
+2009_001070
+2009_001074
+2009_001075
+2009_001078
+2009_001079
+2009_001081
+2009_001082
+2009_001083
+2009_001084
+2009_001085
+2009_001090
+2009_001091
+2009_001094
+2009_001095
+2009_001096
+2009_001097
+2009_001098
+2009_001100
+2009_001102
+2009_001103
+2009_001104
+2009_001105
+2009_001106
+2009_001107
+2009_001108
+2009_001110
+2009_001111
+2009_001113
+2009_001117
+2009_001118
+2009_001120
+2009_001121
+2009_001124
+2009_001126
+2009_001128
+2009_001129
+2009_001133
+2009_001134
+2009_001135
+2009_001137
+2009_001138
+2009_001139
+2009_001140
+2009_001145
+2009_001146
+2009_001147
+2009_001148
+2009_001151
+2009_001152
+2009_001153
+2009_001154
+2009_001155
+2009_001159
+2009_001160
+2009_001163
+2009_001164
+2009_001166
+2009_001172
+2009_001177
+2009_001180
+2009_001181
+2009_001184
+2009_001188
+2009_001190
+2009_001192
+2009_001194
+2009_001195
+2009_001196
+2009_001197
+2009_001198
+2009_001199
+2009_001201
+2009_001203
+2009_001205
+2009_001206
+2009_001207
+2009_001208
+2009_001212
+2009_001215
+2009_001216
+2009_001217
+2009_001221
+2009_001224
+2009_001225
+2009_001227
+2009_001229
+2009_001230
+2009_001236
+2009_001237
+2009_001238
+2009_001240
+2009_001241
+2009_001242
+2009_001243
+2009_001245
+2009_001249
+2009_001251
+2009_001252
+2009_001253
+2009_001254
+2009_001255
+2009_001257
+2009_001259
+2009_001260
+2009_001263
+2009_001264
+2009_001266
+2009_001268
+2009_001270
+2009_001271
+2009_001278
+2009_001279
+2009_001282
+2009_001283
+2009_001285
+2009_001286
+2009_001288
+2009_001289
+2009_001291
+2009_001299
+2009_001300
+2009_001301
+2009_001303
+2009_001305
+2009_001306
+2009_001308
+2009_001309
+2009_001311
+2009_001312
+2009_001313
+2009_001314
+2009_001316
+2009_001319
+2009_001320
+2009_001321
+2009_001322
+2009_001323
+2009_001326
+2009_001327
+2009_001328
+2009_001329
+2009_001332
+2009_001333
+2009_001339
+2009_001343
+2009_001344
+2009_001345
+2009_001348
+2009_001349
+2009_001350
+2009_001354
+2009_001355
+2009_001357
+2009_001359
+2009_001360
+2009_001361
+2009_001363
+2009_001364
+2009_001366
+2009_001367
+2009_001368
+2009_001369
+2009_001370
+2009_001371
+2009_001372
+2009_001374
+2009_001375
+2009_001376
+2009_001384
+2009_001385
+2009_001387
+2009_001388
+2009_001389
+2009_001390
+2009_001391
+2009_001393
+2009_001395
+2009_001397
+2009_001398
+2009_001403
+2009_001406
+2009_001407
+2009_001409
+2009_001411
+2009_001412
+2009_001413
+2009_001414
+2009_001417
+2009_001419
+2009_001422
+2009_001424
+2009_001426
+2009_001427
+2009_001431
+2009_001433
+2009_001434
+2009_001435
+2009_001437
+2009_001440
+2009_001443
+2009_001444
+2009_001446
+2009_001447
+2009_001448
+2009_001449
+2009_001450
+2009_001452
+2009_001453
+2009_001456
+2009_001457
+2009_001462
+2009_001463
+2009_001466
+2009_001468
+2009_001470
+2009_001472
+2009_001474
+2009_001475
+2009_001476
+2009_001479
+2009_001480
+2009_001481
+2009_001484
+2009_001490
+2009_001493
+2009_001494
+2009_001498
+2009_001500
+2009_001501
+2009_001502
+2009_001505
+2009_001507
+2009_001508
+2009_001509
+2009_001514
+2009_001516
+2009_001517
+2009_001518
+2009_001519
+2009_001521
+2009_001522
+2009_001526
+2009_001534
+2009_001535
+2009_001536
+2009_001537
+2009_001538
+2009_001539
+2009_001541
+2009_001542
+2009_001544
+2009_001546
+2009_001549
+2009_001550
+2009_001553
+2009_001554
+2009_001555
+2009_001558
+2009_001562
+2009_001565
+2009_001566
+2009_001567
+2009_001568
+2009_001570
+2009_001575
+2009_001577
+2009_001581
+2009_001585
+2009_001587
+2009_001589
+2009_001590
+2009_001591
+2009_001593
+2009_001594
+2009_001595
+2009_001598
+2009_001602
+2009_001605
+2009_001606
+2009_001607
+2009_001608
+2009_001611
+2009_001612
+2009_001614
+2009_001615
+2009_001617
+2009_001618
+2009_001621
+2009_001623
+2009_001625
+2009_001627
+2009_001631
+2009_001633
+2009_001635
+2009_001636
+2009_001638
+2009_001640
+2009_001642
+2009_001643
+2009_001644
+2009_001645
+2009_001646
+2009_001648
+2009_001651
+2009_001653
+2009_001657
+2009_001660
+2009_001663
+2009_001664
+2009_001667
+2009_001670
+2009_001671
+2009_001673
+2009_001674
+2009_001675
+2009_001676
+2009_001677
+2009_001678
+2009_001682
+2009_001683
+2009_001684
+2009_001687
+2009_001689
+2009_001690
+2009_001693
+2009_001695
+2009_001696
+2009_001699
+2009_001704
+2009_001705
+2009_001706
+2009_001707
+2009_001709
+2009_001713
+2009_001715
+2009_001718
+2009_001719
+2009_001720
+2009_001723
+2009_001724
+2009_001731
+2009_001732
+2009_001733
+2009_001734
+2009_001735
+2009_001738
+2009_001740
+2009_001741
+2009_001743
+2009_001744
+2009_001746
+2009_001747
+2009_001749
+2009_001750
+2009_001751
+2009_001752
+2009_001754
+2009_001755
+2009_001758
+2009_001759
+2009_001764
+2009_001765
+2009_001767
+2009_001768
+2009_001770
+2009_001774
+2009_001775
+2009_001778
+2009_001779
+2009_001780
+2009_001781
+2009_001782
+2009_001783
+2009_001784
+2009_001792
+2009_001794
+2009_001798
+2009_001799
+2009_001800
+2009_001801
+2009_001802
+2009_001804
+2009_001805
+2009_001806
+2009_001807
+2009_001809
+2009_001810
+2009_001811
+2009_001812
+2009_001816
+2009_001817
+2009_001818
+2009_001820
+2009_001822
+2009_001823
+2009_001825
+2009_001826
+2009_001827
+2009_001828
+2009_001830
+2009_001831
+2009_001833
+2009_001835
+2009_001837
+2009_001839
+2009_001840
+2009_001846
+2009_001847
+2009_001848
+2009_001850
+2009_001851
+2009_001852
+2009_001853
+2009_001854
+2009_001856
+2009_001858
+2009_001861
+2009_001864
+2009_001865
+2009_001867
+2009_001868
+2009_001869
+2009_001871
+2009_001873
+2009_001874
+2009_001875
+2009_001881
+2009_001884
+2009_001885
+2009_001888
+2009_001890
+2009_001894
+2009_001897
+2009_001898
+2009_001902
+2009_001904
+2009_001905
+2009_001906
+2009_001907
+2009_001908
+2009_001909
+2009_001910
+2009_001911
+2009_001915
+2009_001916
+2009_001917
+2009_001922
+2009_001926
+2009_001927
+2009_001929
+2009_001931
+2009_001933
+2009_001934
+2009_001937
+2009_001940
+2009_001941
+2009_001945
+2009_001948
+2009_001949
+2009_001952
+2009_001959
+2009_001960
+2009_001961
+2009_001962
+2009_001964
+2009_001965
+2009_001967
+2009_001971
+2009_001972
+2009_001973
+2009_001975
+2009_001976
+2009_001977
+2009_001979
+2009_001980
+2009_001984
+2009_001988
+2009_001990
+2009_001991
+2009_001994
+2009_001997
+2009_001999
+2009_002000
+2009_002001
+2009_002002
+2009_002003
+2009_002008
+2009_002009
+2009_002010
+2009_002011
+2009_002012
+2009_002018
+2009_002019
+2009_002024
+2009_002031
+2009_002035
+2009_002037
+2009_002039
+2009_002040
+2009_002042
+2009_002044
+2009_002046
+2009_002047
+2009_002052
+2009_002053
+2009_002054
+2009_002055
+2009_002056
+2009_002057
+2009_002058
+2009_002060
+2009_002061
+2009_002064
+2009_002066
+2009_002072
+2009_002073
+2009_002077
+2009_002078
+2009_002082
+2009_002083
+2009_002086
+2009_002087
+2009_002088
+2009_002089
+2009_002093
+2009_002094
+2009_002096
+2009_002097
+2009_002098
+2009_002099
+2009_002103
+2009_002104
+2009_002105
+2009_002107
+2009_002110
+2009_002111
+2009_002112
+2009_002116
+2009_002117
+2009_002118
+2009_002119
+2009_002120
+2009_002122
+2009_002123
+2009_002126
+2009_002127
+2009_002128
+2009_002129
+2009_002131
+2009_002133
+2009_002136
+2009_002137
+2009_002139
+2009_002141
+2009_002144
+2009_002145
+2009_002146
+2009_002147
+2009_002149
+2009_002150
+2009_002151
+2009_002152
+2009_002153
+2009_002155
+2009_002164
+2009_002165
+2009_002169
+2009_002171
+2009_002173
+2009_002175
+2009_002176
+2009_002177
+2009_002180
+2009_002182
+2009_002185
+2009_002191
+2009_002192
+2009_002193
+2009_002194
+2009_002197
+2009_002198
+2009_002199
+2009_002202
+2009_002203
+2009_002204
+2009_002205
+2009_002208
+2009_002211
+2009_002212
+2009_002214
+2009_002215
+2009_002216
+2009_002219
+2009_002221
+2009_002222
+2009_002225
+2009_002226
+2009_002228
+2009_002229
+2009_002230
+2009_002231
+2009_002232
+2009_002235
+2009_002236
+2009_002238
+2009_002239
+2009_002240
+2009_002242
+2009_002245
+2009_002252
+2009_002253
+2009_002254
+2009_002256
+2009_002257
+2009_002258
+2009_002259
+2009_002262
+2009_002264
+2009_002265
+2009_002267
+2009_002268
+2009_002271
+2009_002272
+2009_002273
+2009_002274
+2009_002281
+2009_002282
+2009_002285
+2009_002286
+2009_002289
+2009_002291
+2009_002295
+2009_002297
+2009_002298
+2009_002299
+2009_002301
+2009_002302
+2009_002305
+2009_002306
+2009_002308
+2009_002311
+2009_002312
+2009_002314
+2009_002317
+2009_002319
+2009_002320
+2009_002324
+2009_002325
+2009_002326
+2009_002328
+2009_002331
+2009_002333
+2009_002335
+2009_002338
+2009_002339
+2009_002343
+2009_002346
+2009_002348
+2009_002349
+2009_002350
+2009_002352
+2009_002358
+2009_002360
+2009_002362
+2009_002363
+2009_002366
+2009_002370
+2009_002371
+2009_002372
+2009_002373
+2009_002374
+2009_002376
+2009_002377
+2009_002380
+2009_002381
+2009_002382
+2009_002386
+2009_002387
+2009_002388
+2009_002390
+2009_002391
+2009_002393
+2009_002397
+2009_002398
+2009_002399
+2009_002400
+2009_002401
+2009_002404
+2009_002406
+2009_002407
+2009_002408
+2009_002409
+2009_002414
+2009_002415
+2009_002416
+2009_002419
+2009_002420
+2009_002422
+2009_002423
+2009_002424
+2009_002425
+2009_002429
+2009_002431
+2009_002432
+2009_002433
+2009_002434
+2009_002436
+2009_002438
+2009_002439
+2009_002441
+2009_002443
+2009_002444
+2009_002445
+2009_002448
+2009_002449
+2009_002452
+2009_002453
+2009_002456
+2009_002457
+2009_002460
+2009_002464
+2009_002465
+2009_002470
+2009_002471
+2009_002472
+2009_002474
+2009_002475
+2009_002476
+2009_002477
+2009_002487
+2009_002488
+2009_002499
+2009_002500
+2009_002504
+2009_002505
+2009_002506
+2009_002510
+2009_002512
+2009_002514
+2009_002515
+2009_002517
+2009_002518
+2009_002519
+2009_002521
+2009_002522
+2009_002523
+2009_002524
+2009_002525
+2009_002527
+2009_002530
+2009_002531
+2009_002532
+2009_002535
+2009_002536
+2009_002537
+2009_002539
+2009_002542
+2009_002543
+2009_002546
+2009_002549
+2009_002552
+2009_002553
+2009_002556
+2009_002557
+2009_002558
+2009_002559
+2009_002561
+2009_002562
+2009_002563
+2009_002565
+2009_002566
+2009_002567
+2009_002568
+2009_002569
+2009_002570
+2009_002571
+2009_002573
+2009_002577
+2009_002579
+2009_002580
+2009_002584
+2009_002585
+2009_002586
+2009_002588
+2009_002591
+2009_002592
+2009_002594
+2009_002595
+2009_002597
+2009_002599
+2009_002604
+2009_002605
+2009_002607
+2009_002608
+2009_002609
+2009_002611
+2009_002612
+2009_002613
+2009_002614
+2009_002615
+2009_002616
+2009_002618
+2009_002620
+2009_002621
+2009_002624
+2009_002625
+2009_002626
+2009_002628
+2009_002629
+2009_002632
+2009_002634
+2009_002635
+2009_002638
+2009_002645
+2009_002648
+2009_002649
+2009_002651
+2009_002652
+2009_002659
+2009_002662
+2009_002663
+2009_002665
+2009_002667
+2009_002668
+2009_002669
+2009_002670
+2009_002671
+2009_002672
+2009_002673
+2009_002674
+2009_002675
+2009_002676
+2009_002680
+2009_002681
+2009_002683
+2009_002684
+2009_002685
+2009_002687
+2009_002688
+2009_002689
+2009_002695
+2009_002697
+2009_002698
+2009_002703
+2009_002704
+2009_002705
+2009_002708
+2009_002710
+2009_002711
+2009_002712
+2009_002713
+2009_002714
+2009_002715
+2009_002717
+2009_002719
+2009_002725
+2009_002727
+2009_002728
+2009_002732
+2009_002733
+2009_002734
+2009_002739
+2009_002741
+2009_002743
+2009_002744
+2009_002746
+2009_002749
+2009_002750
+2009_002752
+2009_002753
+2009_002754
+2009_002755
+2009_002758
+2009_002759
+2009_002762
+2009_002763
+2009_002764
+2009_002765
+2009_002770
+2009_002771
+2009_002772
+2009_002774
+2009_002777
+2009_002778
+2009_002779
+2009_002780
+2009_002784
+2009_002785
+2009_002789
+2009_002790
+2009_002791
+2009_002792
+2009_002798
+2009_002799
+2009_002800
+2009_002803
+2009_002806
+2009_002807
+2009_002808
+2009_002809
+2009_002813
+2009_002814
+2009_002816
+2009_002817
+2009_002820
+2009_002824
+2009_002827
+2009_002830
+2009_002831
+2009_002833
+2009_002835
+2009_002836
+2009_002837
+2009_002838
+2009_002841
+2009_002842
+2009_002843
+2009_002844
+2009_002845
+2009_002847
+2009_002849
+2009_002850
+2009_002851
+2009_002853
+2009_002855
+2009_002856
+2009_002862
+2009_002865
+2009_002867
+2009_002869
+2009_002872
+2009_002876
+2009_002877
+2009_002879
+2009_002882
+2009_002883
+2009_002885
+2009_002887
+2009_002888
+2009_002890
+2009_002893
+2009_002894
+2009_002897
+2009_002898
+2009_002901
+2009_002902
+2009_002908
+2009_002910
+2009_002912
+2009_002914
+2009_002917
+2009_002918
+2009_002920
+2009_002921
+2009_002925
+2009_002928
+2009_002932
+2009_002933
+2009_002935
+2009_002936
+2009_002937
+2009_002938
+2009_002940
+2009_002941
+2009_002946
+2009_002947
+2009_002952
+2009_002954
+2009_002955
+2009_002957
+2009_002958
+2009_002960
+2009_002961
+2009_002962
+2009_002967
+2009_002970
+2009_002971
+2009_002972
+2009_002975
+2009_002976
+2009_002977
+2009_002978
+2009_002980
+2009_002982
+2009_002983
+2009_002984
+2009_002985
+2009_002986
+2009_002988
+2009_002990
+2009_002993
+2009_002995
+2009_002998
+2009_002999
+2009_003000
+2009_003002
+2009_003003
+2009_003005
+2009_003006
+2009_003007
+2009_003010
+2009_003012
+2009_003013
+2009_003018
+2009_003019
+2009_003020
+2009_003022
+2009_003023
+2009_003031
+2009_003032
+2009_003033
+2009_003034
+2009_003035
+2009_003039
+2009_003042
+2009_003043
+2009_003044
+2009_003052
+2009_003053
+2009_003054
+2009_003056
+2009_003058
+2009_003059
+2009_003063
+2009_003064
+2009_003065
+2009_003066
+2009_003067
+2009_003068
+2009_003070
+2009_003071
+2009_003074
+2009_003075
+2009_003076
+2009_003077
+2009_003078
+2009_003080
+2009_003082
+2009_003083
+2009_003084
+2009_003087
+2009_003088
+2009_003089
+2009_003090
+2009_003091
+2009_003093
+2009_003095
+2009_003097
+2009_003098
+2009_003105
+2009_003107
+2009_003108
+2009_003109
+2009_003110
+2009_003114
+2009_003115
+2009_003116
+2009_003118
+2009_003122
+2009_003123
+2009_003125
+2009_003126
+2009_003127
+2009_003128
+2009_003129
+2009_003130
+2009_003132
+2009_003136
+2009_003138
+2009_003140
+2009_003142
+2009_003143
+2009_003144
+2009_003146
+2009_003147
+2009_003150
+2009_003151
+2009_003153
+2009_003154
+2009_003155
+2009_003156
+2009_003157
+2009_003164
+2009_003165
+2009_003166
+2009_003168
+2009_003172
+2009_003173
+2009_003175
+2009_003181
+2009_003183
+2009_003185
+2009_003187
+2009_003189
+2009_003191
+2009_003193
+2009_003194
+2009_003196
+2009_003198
+2009_003199
+2009_003200
+2009_003201
+2009_003204
+2009_003208
+2009_003209
+2009_003212
+2009_003214
+2009_003217
+2009_003218
+2009_003219
+2009_003222
+2009_003224
+2009_003225
+2009_003229
+2009_003230
+2009_003232
+2009_003233
+2009_003234
+2009_003238
+2009_003241
+2009_003247
+2009_003249
+2009_003251
+2009_003253
+2009_003254
+2009_003255
+2009_003257
+2009_003259
+2009_003261
+2009_003262
+2009_003265
+2009_003266
+2009_003267
+2009_003269
+2009_003271
+2009_003272
+2009_003273
+2009_003276
+2009_003277
+2009_003278
+2009_003282
+2009_003284
+2009_003285
+2009_003288
+2009_003290
+2009_003294
+2009_003297
+2009_003299
+2009_003300
+2009_003301
+2009_003304
+2009_003305
+2009_003309
+2009_003310
+2009_003311
+2009_003312
+2009_003315
+2009_003316
+2009_003317
+2009_003320
+2009_003323
+2009_003326
+2009_003327
+2009_003333
+2009_003338
+2009_003340
+2009_003343
+2009_003345
+2009_003346
+2009_003347
+2009_003348
+2009_003349
+2009_003350
+2009_003351
+2009_003352
+2009_003353
+2009_003360
+2009_003361
+2009_003363
+2009_003365
+2009_003367
+2009_003369
+2009_003372
+2009_003373
+2009_003375
+2009_003376
+2009_003377
+2009_003378
+2009_003379
+2009_003380
+2009_003381
+2009_003383
+2009_003384
+2009_003385
+2009_003386
+2009_003387
+2009_003394
+2009_003395
+2009_003396
+2009_003399
+2009_003400
+2009_003402
+2009_003406
+2009_003407
+2009_003409
+2009_003411
+2009_003415
+2009_003416
+2009_003417
+2009_003419
+2009_003422
+2009_003425
+2009_003430
+2009_003431
+2009_003433
+2009_003436
+2009_003440
+2009_003441
+2009_003443
+2009_003445
+2009_003446
+2009_003447
+2009_003450
+2009_003453
+2009_003454
+2009_003455
+2009_003456
+2009_003457
+2009_003458
+2009_003459
+2009_003460
+2009_003461
+2009_003462
+2009_003466
+2009_003467
+2009_003468
+2009_003469
+2009_003476
+2009_003481
+2009_003482
+2009_003487
+2009_003488
+2009_003489
+2009_003490
+2009_003491
+2009_003492
+2009_003494
+2009_003497
+2009_003498
+2009_003499
+2009_003500
+2009_003504
+2009_003507
+2009_003508
+2009_003509
+2009_003510
+2009_003511
+2009_003513
+2009_003517
+2009_003519
+2009_003520
+2009_003521
+2009_003522
+2009_003523
+2009_003524
+2009_003528
+2009_003530
+2009_003531
+2009_003533
+2009_003534
+2009_003537
+2009_003538
+2009_003539
+2009_003540
+2009_003541
+2009_003542
+2009_003543
+2009_003544
+2009_003545
+2009_003546
+2009_003549
+2009_003551
+2009_003554
+2009_003555
+2009_003560
+2009_003562
+2009_003563
+2009_003564
+2009_003565
+2009_003566
+2009_003569
+2009_003571
+2009_003572
+2009_003576
+2009_003577
+2009_003581
+2009_003583
+2009_003588
+2009_003589
+2009_003592
+2009_003594
+2009_003598
+2009_003600
+2009_003601
+2009_003605
+2009_003606
+2009_003607
+2009_003608
+2009_003609
+2009_003612
+2009_003613
+2009_003614
+2009_003618
+2009_003624
+2009_003626
+2009_003627
+2009_003629
+2009_003633
+2009_003634
+2009_003635
+2009_003636
+2009_003637
+2009_003638
+2009_003639
+2009_003640
+2009_003642
+2009_003644
+2009_003646
+2009_003647
+2009_003650
+2009_003652
+2009_003654
+2009_003655
+2009_003656
+2009_003657
+2009_003660
+2009_003663
+2009_003664
+2009_003666
+2009_003667
+2009_003668
+2009_003669
+2009_003671
+2009_003677
+2009_003679
+2009_003683
+2009_003685
+2009_003686
+2009_003688
+2009_003689
+2009_003690
+2009_003694
+2009_003695
+2009_003696
+2009_003697
+2009_003698
+2009_003702
+2009_003703
+2009_003704
+2009_003705
+2009_003707
+2009_003708
+2009_003709
+2009_003710
+2009_003711
+2009_003713
+2009_003714
+2009_003717
+2009_003718
+2009_003720
+2009_003722
+2009_003725
+2009_003726
+2009_003732
+2009_003734
+2009_003735
+2009_003736
+2009_003738
+2009_003739
+2009_003743
+2009_003747
+2009_003751
+2009_003752
+2009_003753
+2009_003756
+2009_003757
+2009_003758
+2009_003759
+2009_003760
+2009_003765
+2009_003768
+2009_003771
+2009_003773
+2009_003775
+2009_003776
+2009_003781
+2009_003783
+2009_003784
+2009_003785
+2009_003786
+2009_003790
+2009_003793
+2009_003795
+2009_003799
+2009_003800
+2009_003801
+2009_003802
+2009_003804
+2009_003805
+2009_003806
+2009_003808
+2009_003810
+2009_003813
+2009_003814
+2009_003815
+2009_003816
+2009_003818
+2009_003819
+2009_003820
+2009_003821
+2009_003822
+2009_003825
+2009_003827
+2009_003829
+2009_003832
+2009_003835
+2009_003836
+2009_003837
+2009_003838
+2009_003840
+2009_003843
+2009_003846
+2009_003847
+2009_003848
+2009_003849
+2009_003852
+2009_003855
+2009_003857
+2009_003858
+2009_003860
+2009_003863
+2009_003865
+2009_003867
+2009_003870
+2009_003873
+2009_003874
+2009_003879
+2009_003883
+2009_003884
+2009_003888
+2009_003892
+2009_003895
+2009_003896
+2009_003897
+2009_003899
+2009_003900
+2009_003901
+2009_003902
+2009_003903
+2009_003904
+2009_003905
+2009_003908
+2009_003911
+2009_003912
+2009_003913
+2009_003914
+2009_003916
+2009_003920
+2009_003921
+2009_003922
+2009_003928
+2009_003929
+2009_003933
+2009_003936
+2009_003938
+2009_003942
+2009_003944
+2009_003947
+2009_003950
+2009_003951
+2009_003955
+2009_003956
+2009_003958
+2009_003961
+2009_003962
+2009_003965
+2009_003966
+2009_003969
+2009_003971
+2009_003973
+2009_003974
+2009_003975
+2009_003976
+2009_003977
+2009_003982
+2009_003985
+2009_003986
+2009_003991
+2009_003992
+2009_003993
+2009_003994
+2009_003995
+2009_004001
+2009_004002
+2009_004004
+2009_004005
+2009_004007
+2009_004012
+2009_004016
+2009_004018
+2009_004019
+2009_004020
+2009_004021
+2009_004022
+2009_004023
+2009_004025
+2009_004031
+2009_004032
+2009_004033
+2009_004034
+2009_004037
+2009_004038
+2009_004040
+2009_004042
+2009_004043
+2009_004044
+2009_004050
+2009_004051
+2009_004052
+2009_004055
+2009_004058
+2009_004062
+2009_004069
+2009_004070
+2009_004072
+2009_004073
+2009_004074
+2009_004075
+2009_004076
+2009_004078
+2009_004082
+2009_004083
+2009_004084
+2009_004085
+2009_004088
+2009_004091
+2009_004092
+2009_004093
+2009_004094
+2009_004095
+2009_004096
+2009_004099
+2009_004100
+2009_004102
+2009_004103
+2009_004105
+2009_004108
+2009_004109
+2009_004111
+2009_004112
+2009_004113
+2009_004117
+2009_004118
+2009_004121
+2009_004122
+2009_004124
+2009_004125
+2009_004126
+2009_004128
+2009_004129
+2009_004131
+2009_004133
+2009_004134
+2009_004138
+2009_004139
+2009_004140
+2009_004141
+2009_004142
+2009_004148
+2009_004150
+2009_004152
+2009_004153
+2009_004154
+2009_004157
+2009_004159
+2009_004161
+2009_004162
+2009_004163
+2009_004164
+2009_004165
+2009_004166
+2009_004168
+2009_004169
+2009_004170
+2009_004171
+2009_004173
+2009_004174
+2009_004175
+2009_004176
+2009_004177
+2009_004178
+2009_004179
+2009_004180
+2009_004181
+2009_004183
+2009_004186
+2009_004187
+2009_004188
+2009_004191
+2009_004193
+2009_004197
+2009_004199
+2009_004200
+2009_004201
+2009_004202
+2009_004203
+2009_004205
+2009_004207
+2009_004210
+2009_004211
+2009_004212
+2009_004213
+2009_004217
+2009_004218
+2009_004221
+2009_004222
+2009_004224
+2009_004225
+2009_004227
+2009_004228
+2009_004229
+2009_004231
+2009_004232
+2009_004233
+2009_004234
+2009_004241
+2009_004242
+2009_004243
+2009_004244
+2009_004247
+2009_004248
+2009_004249
+2009_004255
+2009_004258
+2009_004261
+2009_004262
+2009_004263
+2009_004264
+2009_004271
+2009_004272
+2009_004273
+2009_004274
+2009_004276
+2009_004277
+2009_004278
+2009_004279
+2009_004283
+2009_004284
+2009_004285
+2009_004289
+2009_004290
+2009_004291
+2009_004295
+2009_004298
+2009_004300
+2009_004301
+2009_004303
+2009_004307
+2009_004308
+2009_004309
+2009_004312
+2009_004315
+2009_004316
+2009_004317
+2009_004319
+2009_004322
+2009_004323
+2009_004324
+2009_004327
+2009_004328
+2009_004329
+2009_004332
+2009_004334
+2009_004336
+2009_004338
+2009_004340
+2009_004341
+2009_004346
+2009_004347
+2009_004350
+2009_004351
+2009_004357
+2009_004358
+2009_004359
+2009_004361
+2009_004364
+2009_004366
+2009_004368
+2009_004369
+2009_004370
+2009_004371
+2009_004374
+2009_004375
+2009_004377
+2009_004382
+2009_004383
+2009_004390
+2009_004392
+2009_004394
+2009_004397
+2009_004399
+2009_004402
+2009_004403
+2009_004404
+2009_004406
+2009_004409
+2009_004410
+2009_004411
+2009_004414
+2009_004417
+2009_004419
+2009_004424
+2009_004425
+2009_004426
+2009_004429
+2009_004432
+2009_004434
+2009_004435
+2009_004436
+2009_004438
+2009_004440
+2009_004442
+2009_004444
+2009_004445
+2009_004446
+2009_004448
+2009_004449
+2009_004451
+2009_004452
+2009_004453
+2009_004454
+2009_004455
+2009_004456
+2009_004457
+2009_004464
+2009_004465
+2009_004468
+2009_004471
+2009_004475
+2009_004477
+2009_004478
+2009_004479
+2009_004483
+2009_004486
+2009_004492
+2009_004494
+2009_004496
+2009_004497
+2009_004499
+2009_004501
+2009_004502
+2009_004503
+2009_004504
+2009_004507
+2009_004508
+2009_004509
+2009_004511
+2009_004513
+2009_004514
+2009_004518
+2009_004519
+2009_004524
+2009_004525
+2009_004527
+2009_004529
+2009_004530
+2009_004532
+2009_004535
+2009_004536
+2009_004537
+2009_004539
+2009_004540
+2009_004542
+2009_004543
+2009_004545
+2009_004547
+2009_004548
+2009_004551
+2009_004552
+2009_004554
+2009_004556
+2009_004557
+2009_004559
+2009_004560
+2009_004561
+2009_004562
+2009_004565
+2009_004567
+2009_004568
+2009_004570
+2009_004571
+2009_004572
+2009_004579
+2009_004580
+2009_004581
+2009_004582
+2009_004587
+2009_004588
+2009_004590
+2009_004592
+2009_004593
+2009_004594
+2009_004598
+2009_004601
+2009_004606
+2009_004607
+2009_004614
+2009_004616
+2009_004619
+2009_004620
+2009_004623
+2009_004624
+2009_004625
+2009_004626
+2009_004628
+2009_004629
+2009_004630
+2009_004631
+2009_004634
+2009_004635
+2009_004639
+2009_004642
+2009_004643
+2009_004645
+2009_004647
+2009_004648
+2009_004651
+2009_004652
+2009_004653
+2009_004655
+2009_004656
+2009_004661
+2009_004662
+2009_004664
+2009_004667
+2009_004669
+2009_004670
+2009_004671
+2009_004674
+2009_004677
+2009_004679
+2009_004681
+2009_004683
+2009_004684
+2009_004686
+2009_004687
+2009_004688
+2009_004694
+2009_004697
+2009_004701
+2009_004705
+2009_004706
+2009_004708
+2009_004709
+2009_004710
+2009_004713
+2009_004716
+2009_004718
+2009_004719
+2009_004720
+2009_004721
+2009_004723
+2009_004728
+2009_004730
+2009_004731
+2009_004732
+2009_004734
+2009_004737
+2009_004738
+2009_004744
+2009_004745
+2009_004746
+2009_004748
+2009_004749
+2009_004754
+2009_004756
+2009_004758
+2009_004759
+2009_004760
+2009_004761
+2009_004763
+2009_004764
+2009_004765
+2009_004766
+2009_004768
+2009_004769
+2009_004771
+2009_004772
+2009_004779
+2009_004780
+2009_004781
+2009_004782
+2009_004784
+2009_004786
+2009_004787
+2009_004789
+2009_004790
+2009_004794
+2009_004796
+2009_004797
+2009_004798
+2009_004799
+2009_004801
+2009_004804
+2009_004805
+2009_004806
+2009_004812
+2009_004813
+2009_004815
+2009_004817
+2009_004820
+2009_004822
+2009_004823
+2009_004824
+2009_004828
+2009_004829
+2009_004830
+2009_004831
+2009_004834
+2009_004836
+2009_004839
+2009_004841
+2009_004845
+2009_004846
+2009_004847
+2009_004848
+2009_004849
+2009_004855
+2009_004856
+2009_004857
+2009_004858
+2009_004859
+2009_004865
+2009_004867
+2009_004868
+2009_004869
+2009_004871
+2009_004872
+2009_004874
+2009_004876
+2009_004877
+2009_004880
+2009_004882
+2009_004885
+2009_004886
+2009_004887
+2009_004888
+2009_004889
+2009_004890
+2009_004895
+2009_004897
+2009_004898
+2009_004899
+2009_004901
+2009_004902
+2009_004903
+2009_004904
+2009_004905
+2009_004907
+2009_004913
+2009_004914
+2009_004917
+2009_004919
+2009_004921
+2009_004922
+2009_004926
+2009_004929
+2009_004930
+2009_004933
+2009_004934
+2009_004939
+2009_004940
+2009_004942
+2009_004943
+2009_004944
+2009_004945
+2009_004946
+2009_004947
+2009_004953
+2009_004956
+2009_004958
+2009_004959
+2009_004961
+2009_004962
+2009_004965
+2009_004969
+2009_004971
+2009_004972
+2009_004974
+2009_004975
+2009_004977
+2009_004979
+2009_004980
+2009_004982
+2009_004983
+2009_004984
+2009_004986
+2009_004987
+2009_004988
+2009_004990
+2009_004993
+2009_004994
+2009_004996
+2009_004999
+2009_005000
+2009_005001
+2009_005005
+2009_005006
+2009_005008
+2009_005015
+2009_005016
+2009_005019
+2009_005024
+2009_005025
+2009_005030
+2009_005031
+2009_005033
+2009_005035
+2009_005036
+2009_005037
+2009_005038
+2009_005040
+2009_005042
+2009_005044
+2009_005045
+2009_005051
+2009_005055
+2009_005056
+2009_005057
+2009_005060
+2009_005061
+2009_005062
+2009_005064
+2009_005068
+2009_005069
+2009_005070
+2009_005073
+2009_005075
+2009_005076
+2009_005078
+2009_005079
+2009_005080
+2009_005081
+2009_005082
+2009_005083
+2009_005084
+2009_005085
+2009_005086
+2009_005087
+2009_005089
+2009_005094
+2009_005095
+2009_005098
+2009_005102
+2009_005103
+2009_005104
+2009_005107
+2009_005111
+2009_005114
+2009_005118
+2009_005119
+2009_005120
+2009_005126
+2009_005127
+2009_005128
+2009_005129
+2009_005130
+2009_005131
+2009_005133
+2009_005137
+2009_005140
+2009_005141
+2009_005142
+2009_005144
+2009_005145
+2009_005147
+2009_005148
+2009_005149
+2009_005150
+2009_005152
+2009_005153
+2009_005154
+2009_005155
+2009_005156
+2009_005158
+2009_005160
+2009_005161
+2009_005162
+2009_005163
+2009_005165
+2009_005168
+2009_005170
+2009_005171
+2009_005172
+2009_005177
+2009_005178
+2009_005181
+2009_005183
+2009_005185
+2009_005189
+2009_005190
+2009_005191
+2009_005193
+2009_005194
+2009_005198
+2009_005201
+2009_005202
+2009_005203
+2009_005204
+2009_005205
+2009_005210
+2009_005211
+2009_005215
+2009_005216
+2009_005217
+2009_005218
+2009_005219
+2009_005220
+2009_005221
+2009_005222
+2009_005225
+2009_005229
+2009_005231
+2009_005232
+2009_005234
+2009_005236
+2009_005239
+2009_005240
+2009_005242
+2009_005246
+2009_005247
+2009_005251
+2009_005256
+2009_005257
+2009_005260
+2009_005262
+2009_005263
+2009_005265
+2009_005267
+2009_005268
+2009_005269
+2009_005272
+2009_005278
+2009_005279
+2009_005282
+2009_005286
+2009_005287
+2009_005288
+2009_005292
+2009_005293
+2009_005294
+2009_005297
+2009_005299
+2009_005300
+2009_005302
+2009_005303
+2009_005307
+2009_005308
+2009_005309
+2009_005310
+2009_005311
+2010_000001
+2010_000002
+2010_000003
+2010_000009
+2010_000014
+2010_000015
+2010_000018
+2010_000020
+2010_000023
+2010_000024
+2010_000026
+2010_000027
+2010_000031
+2010_000033
+2010_000035
+2010_000036
+2010_000038
+2010_000043
+2010_000045
+2010_000048
+2010_000050
+2010_000052
+2010_000053
+2010_000054
+2010_000055
+2010_000056
+2010_000061
+2010_000063
+2010_000065
+2010_000067
+2010_000069
+2010_000071
+2010_000072
+2010_000073
+2010_000074
+2010_000075
+2010_000076
+2010_000079
+2010_000080
+2010_000082
+2010_000083
+2010_000084
+2010_000085
+2010_000087
+2010_000088
+2010_000089
+2010_000090
+2010_000091
+2010_000095
+2010_000097
+2010_000098
+2010_000099
+2010_000103
+2010_000109
+2010_000110
+2010_000111
+2010_000113
+2010_000114
+2010_000117
+2010_000118
+2010_000120
+2010_000124
+2010_000127
+2010_000131
+2010_000132
+2010_000133
+2010_000136
+2010_000137
+2010_000138
+2010_000139
+2010_000140
+2010_000141
+2010_000145
+2010_000148
+2010_000151
+2010_000152
+2010_000157
+2010_000159
+2010_000160
+2010_000162
+2010_000163
+2010_000165
+2010_000169
+2010_000170
+2010_000172
+2010_000174
+2010_000175
+2010_000177
+2010_000178
+2010_000182
+2010_000183
+2010_000184
+2010_000187
+2010_000189
+2010_000190
+2010_000193
+2010_000194
+2010_000195
+2010_000196
+2010_000197
+2010_000198
+2010_000199
+2010_000202
+2010_000203
+2010_000204
+2010_000209
+2010_000211
+2010_000213
+2010_000216
+2010_000218
+2010_000222
+2010_000224
+2010_000227
+2010_000229
+2010_000233
+2010_000234
+2010_000238
+2010_000241
+2010_000244
+2010_000245
+2010_000246
+2010_000247
+2010_000248
+2010_000249
+2010_000250
+2010_000254
+2010_000255
+2010_000256
+2010_000260
+2010_000261
+2010_000262
+2010_000263
+2010_000264
+2010_000266
+2010_000269
+2010_000270
+2010_000272
+2010_000273
+2010_000276
+2010_000279
+2010_000283
+2010_000284
+2010_000285
+2010_000286
+2010_000291
+2010_000293
+2010_000295
+2010_000296
+2010_000299
+2010_000302
+2010_000303
+2010_000307
+2010_000308
+2010_000309
+2010_000310
+2010_000312
+2010_000313
+2010_000317
+2010_000318
+2010_000320
+2010_000321
+2010_000323
+2010_000324
+2010_000325
+2010_000327
+2010_000329
+2010_000330
+2010_000335
+2010_000336
+2010_000337
+2010_000342
+2010_000344
+2010_000347
+2010_000349
+2010_000351
+2010_000352
+2010_000356
+2010_000358
+2010_000361
+2010_000362
+2010_000370
+2010_000371
+2010_000372
+2010_000374
+2010_000375
+2010_000376
+2010_000377
+2010_000379
+2010_000381
+2010_000382
+2010_000384
+2010_000386
+2010_000388
+2010_000389
+2010_000390
+2010_000392
+2010_000393
+2010_000394
+2010_000395
+2010_000399
+2010_000401
+2010_000404
+2010_000406
+2010_000409
+2010_000413
+2010_000415
+2010_000418
+2010_000419
+2010_000420
+2010_000422
+2010_000426
+2010_000427
+2010_000431
+2010_000432
+2010_000433
+2010_000435
+2010_000436
+2010_000437
+2010_000439
+2010_000442
+2010_000444
+2010_000446
+2010_000447
+2010_000448
+2010_000449
+2010_000453
+2010_000456
+2010_000458
+2010_000459
+2010_000461
+2010_000462
+2010_000463
+2010_000465
+2010_000466
+2010_000468
+2010_000469
+2010_000470
+2010_000473
+2010_000474
+2010_000475
+2010_000477
+2010_000480
+2010_000483
+2010_000484
+2010_000485
+2010_000488
+2010_000490
+2010_000492
+2010_000493
+2010_000495
+2010_000497
+2010_000498
+2010_000500
+2010_000502
+2010_000503
+2010_000506
+2010_000508
+2010_000510
+2010_000511
+2010_000513
+2010_000515
+2010_000519
+2010_000522
+2010_000524
+2010_000526
+2010_000527
+2010_000530
+2010_000534
+2010_000536
+2010_000537
+2010_000538
+2010_000541
+2010_000545
+2010_000547
+2010_000548
+2010_000549
+2010_000552
+2010_000553
+2010_000556
+2010_000557
+2010_000559
+2010_000561
+2010_000562
+2010_000564
+2010_000567
+2010_000568
+2010_000571
+2010_000572
+2010_000573
+2010_000574
+2010_000576
+2010_000577
+2010_000578
+2010_000581
+2010_000582
+2010_000583
+2010_000586
+2010_000588
+2010_000590
+2010_000591
+2010_000601
+2010_000602
+2010_000603
+2010_000604
+2010_000608
+2010_000613
+2010_000616
+2010_000617
+2010_000621
+2010_000622
+2010_000624
+2010_000626
+2010_000628
+2010_000630
+2010_000632
+2010_000633
+2010_000635
+2010_000639
+2010_000641
+2010_000644
+2010_000645
+2010_000646
+2010_000647
+2010_000648
+2010_000651
+2010_000655
+2010_000658
+2010_000661
+2010_000664
+2010_000665
+2010_000666
+2010_000667
+2010_000669
+2010_000671
+2010_000674
+2010_000675
+2010_000678
+2010_000679
+2010_000681
+2010_000682
+2010_000683
+2010_000685
+2010_000687
+2010_000688
+2010_000689
+2010_000690
+2010_000691
+2010_000692
+2010_000694
+2010_000695
+2010_000697
+2010_000702
+2010_000705
+2010_000707
+2010_000710
+2010_000711
+2010_000712
+2010_000715
+2010_000716
+2010_000717
+2010_000721
+2010_000722
+2010_000723
+2010_000724
+2010_000726
+2010_000727
+2010_000729
+2010_000731
+2010_000735
+2010_000737
+2010_000738
+2010_000739
+2010_000740
+2010_000743
+2010_000744
+2010_000746
+2010_000747
+2010_000748
+2010_000749
+2010_000750
+2010_000754
+2010_000759
+2010_000760
+2010_000761
+2010_000764
+2010_000765
+2010_000769
+2010_000770
+2010_000771
+2010_000772
+2010_000773
+2010_000778
+2010_000782
+2010_000785
+2010_000786
+2010_000787
+2010_000788
+2010_000791
+2010_000792
+2010_000797
+2010_000799
+2010_000800
+2010_000802
+2010_000803
+2010_000805
+2010_000806
+2010_000807
+2010_000808
+2010_000810
+2010_000811
+2010_000814
+2010_000815
+2010_000821
+2010_000822
+2010_000827
+2010_000828
+2010_000829
+2010_000830
+2010_000831
+2010_000836
+2010_000837
+2010_000838
+2010_000842
+2010_000846
+2010_000847
+2010_000849
+2010_000855
+2010_000857
+2010_000860
+2010_000862
+2010_000863
+2010_000865
+2010_000866
+2010_000870
+2010_000871
+2010_000872
+2010_000874
+2010_000875
+2010_000876
+2010_000879
+2010_000883
+2010_000885
+2010_000887
+2010_000889
+2010_000891
+2010_000893
+2010_000897
+2010_000898
+2010_000899
+2010_000904
+2010_000906
+2010_000907
+2010_000908
+2010_000910
+2010_000912
+2010_000914
+2010_000915
+2010_000918
+2010_000920
+2010_000922
+2010_000923
+2010_000926
+2010_000927
+2010_000928
+2010_000929
+2010_000931
+2010_000938
+2010_000939
+2010_000941
+2010_000942
+2010_000944
+2010_000945
+2010_000947
+2010_000948
+2010_000952
+2010_000954
+2010_000955
+2010_000956
+2010_000959
+2010_000961
+2010_000968
+2010_000970
+2010_000971
+2010_000973
+2010_000974
+2010_000975
+2010_000978
+2010_000979
+2010_000981
+2010_000983
+2010_000984
+2010_000986
+2010_000989
+2010_000991
+2010_000993
+2010_000994
+2010_000995
+2010_000996
+2010_001000
+2010_001002
+2010_001006
+2010_001008
+2010_001009
+2010_001010
+2010_001011
+2010_001012
+2010_001013
+2010_001016
+2010_001017
+2010_001020
+2010_001021
+2010_001023
+2010_001024
+2010_001025
+2010_001030
+2010_001032
+2010_001036
+2010_001039
+2010_001042
+2010_001043
+2010_001044
+2010_001049
+2010_001051
+2010_001052
+2010_001054
+2010_001057
+2010_001061
+2010_001063
+2010_001066
+2010_001069
+2010_001070
+2010_001074
+2010_001076
+2010_001077
+2010_001079
+2010_001080
+2010_001082
+2010_001085
+2010_001087
+2010_001089
+2010_001092
+2010_001094
+2010_001098
+2010_001099
+2010_001100
+2010_001103
+2010_001104
+2010_001105
+2010_001106
+2010_001107
+2010_001109
+2010_001110
+2010_001111
+2010_001112
+2010_001113
+2010_001117
+2010_001118
+2010_001119
+2010_001120
+2010_001121
+2010_001123
+2010_001124
+2010_001125
+2010_001126
+2010_001127
+2010_001130
+2010_001131
+2010_001134
+2010_001139
+2010_001140
+2010_001142
+2010_001143
+2010_001147
+2010_001148
+2010_001149
+2010_001151
+2010_001152
+2010_001154
+2010_001158
+2010_001159
+2010_001160
+2010_001163
+2010_001164
+2010_001172
+2010_001174
+2010_001175
+2010_001177
+2010_001179
+2010_001181
+2010_001183
+2010_001184
+2010_001185
+2010_001188
+2010_001189
+2010_001192
+2010_001193
+2010_001195
+2010_001199
+2010_001201
+2010_001204
+2010_001205
+2010_001206
+2010_001210
+2010_001211
+2010_001212
+2010_001214
+2010_001215
+2010_001216
+2010_001218
+2010_001219
+2010_001220
+2010_001224
+2010_001225
+2010_001229
+2010_001234
+2010_001237
+2010_001240
+2010_001241
+2010_001242
+2010_001245
+2010_001246
+2010_001247
+2010_001250
+2010_001251
+2010_001253
+2010_001254
+2010_001256
+2010_001257
+2010_001261
+2010_001263
+2010_001264
+2010_001270
+2010_001271
+2010_001272
+2010_001273
+2010_001274
+2010_001275
+2010_001277
+2010_001279
+2010_001282
+2010_001286
+2010_001287
+2010_001288
+2010_001289
+2010_001291
+2010_001292
+2010_001293
+2010_001294
+2010_001299
+2010_001301
+2010_001305
+2010_001310
+2010_001311
+2010_001312
+2010_001313
+2010_001315
+2010_001317
+2010_001320
+2010_001321
+2010_001325
+2010_001326
+2010_001327
+2010_001328
+2010_001329
+2010_001331
+2010_001333
+2010_001337
+2010_001338
+2010_001339
+2010_001343
+2010_001344
+2010_001347
+2010_001351
+2010_001355
+2010_001356
+2010_001357
+2010_001360
+2010_001361
+2010_001363
+2010_001364
+2010_001366
+2010_001367
+2010_001370
+2010_001372
+2010_001374
+2010_001376
+2010_001382
+2010_001383
+2010_001385
+2010_001386
+2010_001390
+2010_001394
+2010_001395
+2010_001397
+2010_001399
+2010_001401
+2010_001402
+2010_001403
+2010_001405
+2010_001406
+2010_001407
+2010_001408
+2010_001410
+2010_001411
+2010_001412
+2010_001413
+2010_001417
+2010_001418
+2010_001421
+2010_001422
+2010_001425
+2010_001426
+2010_001430
+2010_001431
+2010_001432
+2010_001433
+2010_001434
+2010_001435
+2010_001439
+2010_001441
+2010_001448
+2010_001449
+2010_001450
+2010_001451
+2010_001452
+2010_001453
+2010_001455
+2010_001456
+2010_001457
+2010_001458
+2010_001461
+2010_001463
+2010_001464
+2010_001465
+2010_001468
+2010_001472
+2010_001473
+2010_001478
+2010_001479
+2010_001480
+2010_001481
+2010_001486
+2010_001487
+2010_001489
+2010_001497
+2010_001499
+2010_001501
+2010_001502
+2010_001503
+2010_001505
+2010_001511
+2010_001514
+2010_001515
+2010_001516
+2010_001518
+2010_001520
+2010_001522
+2010_001525
+2010_001528
+2010_001529
+2010_001533
+2010_001534
+2010_001535
+2010_001536
+2010_001537
+2010_001539
+2010_001540
+2010_001543
+2010_001544
+2010_001547
+2010_001548
+2010_001550
+2010_001551
+2010_001552
+2010_001553
+2010_001555
+2010_001557
+2010_001560
+2010_001561
+2010_001562
+2010_001563
+2010_001569
+2010_001571
+2010_001572
+2010_001574
+2010_001576
+2010_001577
+2010_001579
+2010_001580
+2010_001583
+2010_001584
+2010_001586
+2010_001587
+2010_001590
+2010_001592
+2010_001594
+2010_001595
+2010_001596
+2010_001599
+2010_001601
+2010_001602
+2010_001603
+2010_001606
+2010_001607
+2010_001608
+2010_001614
+2010_001618
+2010_001619
+2010_001625
+2010_001626
+2010_001630
+2010_001633
+2010_001635
+2010_001636
+2010_001637
+2010_001638
+2010_001640
+2010_001644
+2010_001645
+2010_001646
+2010_001647
+2010_001649
+2010_001650
+2010_001652
+2010_001656
+2010_001659
+2010_001660
+2010_001665
+2010_001668
+2010_001669
+2010_001671
+2010_001674
+2010_001675
+2010_001676
+2010_001679
+2010_001680
+2010_001682
+2010_001685
+2010_001687
+2010_001689
+2010_001690
+2010_001691
+2010_001692
+2010_001694
+2010_001697
+2010_001698
+2010_001699
+2010_001700
+2010_001705
+2010_001706
+2010_001709
+2010_001710
+2010_001712
+2010_001715
+2010_001717
+2010_001718
+2010_001719
+2010_001720
+2010_001726
+2010_001729
+2010_001731
+2010_001732
+2010_001734
+2010_001737
+2010_001739
+2010_001743
+2010_001744
+2010_001746
+2010_001747
+2010_001748
+2010_001749
+2010_001752
+2010_001753
+2010_001754
+2010_001756
+2010_001757
+2010_001759
+2010_001760
+2010_001762
+2010_001763
+2010_001767
+2010_001768
+2010_001770
+2010_001771
+2010_001773
+2010_001776
+2010_001777
+2010_001780
+2010_001783
+2010_001784
+2010_001785
+2010_001787
+2010_001788
+2010_001794
+2010_001795
+2010_001796
+2010_001797
+2010_001801
+2010_001803
+2010_001806
+2010_001807
+2010_001808
+2010_001810
+2010_001814
+2010_001817
+2010_001819
+2010_001820
+2010_001821
+2010_001823
+2010_001827
+2010_001828
+2010_001829
+2010_001830
+2010_001836
+2010_001837
+2010_001838
+2010_001841
+2010_001842
+2010_001843
+2010_001845
+2010_001846
+2010_001849
+2010_001850
+2010_001851
+2010_001852
+2010_001853
+2010_001856
+2010_001857
+2010_001858
+2010_001860
+2010_001862
+2010_001863
+2010_001864
+2010_001868
+2010_001869
+2010_001870
+2010_001877
+2010_001881
+2010_001884
+2010_001885
+2010_001891
+2010_001892
+2010_001893
+2010_001896
+2010_001899
+2010_001904
+2010_001907
+2010_001908
+2010_001911
+2010_001913
+2010_001916
+2010_001918
+2010_001919
+2010_001921
+2010_001922
+2010_001923
+2010_001924
+2010_001927
+2010_001929
+2010_001931
+2010_001933
+2010_001934
+2010_001937
+2010_001938
+2010_001939
+2010_001940
+2010_001941
+2010_001944
+2010_001948
+2010_001950
+2010_001951
+2010_001954
+2010_001956
+2010_001957
+2010_001960
+2010_001962
+2010_001966
+2010_001967
+2010_001968
+2010_001970
+2010_001973
+2010_001974
+2010_001976
+2010_001978
+2010_001979
+2010_001980
+2010_001981
+2010_001982
+2010_001986
+2010_001987
+2010_001988
+2010_001992
+2010_001993
+2010_001994
+2010_001995
+2010_001998
+2010_002000
+2010_002002
+2010_002005
+2010_002006
+2010_002015
+2010_002017
+2010_002018
+2010_002019
+2010_002020
+2010_002022
+2010_002023
+2010_002025
+2010_002026
+2010_002029
+2010_002030
+2010_002032
+2010_002037
+2010_002039
+2010_002040
+2010_002041
+2010_002042
+2010_002044
+2010_002045
+2010_002046
+2010_002047
+2010_002048
+2010_002050
+2010_002054
+2010_002055
+2010_002057
+2010_002058
+2010_002060
+2010_002065
+2010_002067
+2010_002068
+2010_002070
+2010_002073
+2010_002080
+2010_002085
+2010_002086
+2010_002089
+2010_002094
+2010_002095
+2010_002096
+2010_002097
+2010_002098
+2010_002100
+2010_002102
+2010_002104
+2010_002105
+2010_002106
+2010_002107
+2010_002113
+2010_002117
+2010_002118
+2010_002121
+2010_002124
+2010_002127
+2010_002128
+2010_002129
+2010_002130
+2010_002132
+2010_002133
+2010_002136
+2010_002137
+2010_002138
+2010_002139
+2010_002141
+2010_002142
+2010_002143
+2010_002146
+2010_002147
+2010_002149
+2010_002150
+2010_002152
+2010_002154
+2010_002161
+2010_002166
+2010_002167
+2010_002168
+2010_002172
+2010_002175
+2010_002176
+2010_002177
+2010_002179
+2010_002180
+2010_002181
+2010_002182
+2010_002183
+2010_002185
+2010_002187
+2010_002191
+2010_002192
+2010_002193
+2010_002194
+2010_002195
+2010_002199
+2010_002200
+2010_002203
+2010_002204
+2010_002207
+2010_002208
+2010_002211
+2010_002213
+2010_002215
+2010_002216
+2010_002218
+2010_002219
+2010_002220
+2010_002221
+2010_002223
+2010_002224
+2010_002226
+2010_002227
+2010_002228
+2010_002229
+2010_002232
+2010_002236
+2010_002242
+2010_002243
+2010_002244
+2010_002245
+2010_002247
+2010_002248
+2010_002251
+2010_002254
+2010_002255
+2010_002261
+2010_002263
+2010_002267
+2010_002269
+2010_002271
+2010_002274
+2010_002276
+2010_002278
+2010_002279
+2010_002283
+2010_002286
+2010_002287
+2010_002289
+2010_002294
+2010_002295
+2010_002299
+2010_002301
+2010_002303
+2010_002305
+2010_002307
+2010_002309
+2010_002310
+2010_002312
+2010_002313
+2010_002315
+2010_002316
+2010_002318
+2010_002319
+2010_002320
+2010_002321
+2010_002326
+2010_002327
+2010_002332
+2010_002333
+2010_002336
+2010_002337
+2010_002338
+2010_002340
+2010_002346
+2010_002348
+2010_002349
+2010_002353
+2010_002354
+2010_002356
+2010_002357
+2010_002361
+2010_002363
+2010_002364
+2010_002365
+2010_002366
+2010_002368
+2010_002369
+2010_002370
+2010_002371
+2010_002372
+2010_002373
+2010_002374
+2010_002378
+2010_002379
+2010_002382
+2010_002383
+2010_002387
+2010_002388
+2010_002390
+2010_002391
+2010_002392
+2010_002393
+2010_002396
+2010_002398
+2010_002399
+2010_002400
+2010_002402
+2010_002405
+2010_002406
+2010_002408
+2010_002409
+2010_002410
+2010_002413
+2010_002418
+2010_002420
+2010_002422
+2010_002424
+2010_002425
+2010_002427
+2010_002429
+2010_002431
+2010_002435
+2010_002436
+2010_002438
+2010_002439
+2010_002440
+2010_002445
+2010_002446
+2010_002448
+2010_002449
+2010_002450
+2010_002452
+2010_002455
+2010_002456
+2010_002457
+2010_002458
+2010_002459
+2010_002460
+2010_002461
+2010_002462
+2010_002468
+2010_002469
+2010_002472
+2010_002475
+2010_002479
+2010_002480
+2010_002482
+2010_002484
+2010_002485
+2010_002487
+2010_002492
+2010_002496
+2010_002497
+2010_002498
+2010_002499
+2010_002501
+2010_002504
+2010_002507
+2010_002509
+2010_002510
+2010_002512
+2010_002513
+2010_002516
+2010_002518
+2010_002520
+2010_002526
+2010_002527
+2010_002529
+2010_002531
+2010_002532
+2010_002533
+2010_002534
+2010_002536
+2010_002537
+2010_002538
+2010_002539
+2010_002542
+2010_002543
+2010_002546
+2010_002547
+2010_002551
+2010_002552
+2010_002553
+2010_002556
+2010_002561
+2010_002562
+2010_002565
+2010_002567
+2010_002569
+2010_002570
+2010_002573
+2010_002575
+2010_002577
+2010_002578
+2010_002579
+2010_002580
+2010_002582
+2010_002583
+2010_002586
+2010_002587
+2010_002589
+2010_002592
+2010_002594
+2010_002597
+2010_002598
+2010_002601
+2010_002602
+2010_002603
+2010_002605
+2010_002614
+2010_002615
+2010_002616
+2010_002618
+2010_002620
+2010_002621
+2010_002623
+2010_002624
+2010_002625
+2010_002626
+2010_002628
+2010_002629
+2010_002631
+2010_002632
+2010_002638
+2010_002639
+2010_002642
+2010_002644
+2010_002645
+2010_002647
+2010_002652
+2010_002653
+2010_002654
+2010_002656
+2010_002659
+2010_002660
+2010_002661
+2010_002662
+2010_002665
+2010_002666
+2010_002667
+2010_002668
+2010_002674
+2010_002675
+2010_002676
+2010_002678
+2010_002679
+2010_002682
+2010_002684
+2010_002686
+2010_002688
+2010_002691
+2010_002692
+2010_002693
+2010_002695
+2010_002696
+2010_002697
+2010_002701
+2010_002702
+2010_002704
+2010_002705
+2010_002708
+2010_002710
+2010_002713
+2010_002714
+2010_002716
+2010_002720
+2010_002721
+2010_002722
+2010_002723
+2010_002725
+2010_002728
+2010_002729
+2010_002733
+2010_002734
+2010_002736
+2010_002737
+2010_002740
+2010_002741
+2010_002742
+2010_002746
+2010_002747
+2010_002750
+2010_002752
+2010_002754
+2010_002758
+2010_002759
+2010_002760
+2010_002763
+2010_002767
+2010_002770
+2010_002771
+2010_002772
+2010_002774
+2010_002775
+2010_002778
+2010_002779
+2010_002780
+2010_002781
+2010_002783
+2010_002786
+2010_002789
+2010_002790
+2010_002791
+2010_002792
+2010_002793
+2010_002794
+2010_002797
+2010_002801
+2010_002803
+2010_002805
+2010_002807
+2010_002808
+2010_002811
+2010_002813
+2010_002814
+2010_002815
+2010_002816
+2010_002817
+2010_002820
+2010_002821
+2010_002822
+2010_002824
+2010_002827
+2010_002830
+2010_002831
+2010_002834
+2010_002838
+2010_002839
+2010_002840
+2010_002841
+2010_002842
+2010_002843
+2010_002844
+2010_002845
+2010_002851
+2010_002853
+2010_002854
+2010_002855
+2010_002856
+2010_002857
+2010_002858
+2010_002860
+2010_002864
+2010_002865
+2010_002868
+2010_002870
+2010_002871
+2010_002873
+2010_002876
+2010_002877
+2010_002879
+2010_002880
+2010_002881
+2010_002884
+2010_002887
+2010_002891
+2010_002892
+2010_002896
+2010_002899
+2010_002900
+2010_002901
+2010_002902
+2010_002903
+2010_002905
+2010_002907
+2010_002909
+2010_002914
+2010_002915
+2010_002917
+2010_002921
+2010_002924
+2010_002927
+2010_002929
+2010_002930
+2010_002931
+2010_002935
+2010_002937
+2010_002938
+2010_002939
+2010_002940
+2010_002941
+2010_002946
+2010_002947
+2010_002948
+2010_002954
+2010_002955
+2010_002956
+2010_002958
+2010_002960
+2010_002962
+2010_002963
+2010_002965
+2010_002972
+2010_002973
+2010_002976
+2010_002978
+2010_002979
+2010_002980
+2010_002982
+2010_002985
+2010_002987
+2010_002988
+2010_002990
+2010_002991
+2010_002993
+2010_002995
+2010_003002
+2010_003003
+2010_003007
+2010_003010
+2010_003011
+2010_003013
+2010_003014
+2010_003015
+2010_003016
+2010_003017
+2010_003019
+2010_003024
+2010_003025
+2010_003027
+2010_003028
+2010_003032
+2010_003034
+2010_003035
+2010_003037
+2010_003040
+2010_003043
+2010_003044
+2010_003047
+2010_003050
+2010_003051
+2010_003053
+2010_003054
+2010_003055
+2010_003056
+2010_003057
+2010_003060
+2010_003062
+2010_003067
+2010_003071
+2010_003072
+2010_003074
+2010_003077
+2010_003078
+2010_003081
+2010_003082
+2010_003084
+2010_003086
+2010_003088
+2010_003091
+2010_003092
+2010_003093
+2010_003094
+2010_003097
+2010_003098
+2010_003101
+2010_003102
+2010_003103
+2010_003106
+2010_003107
+2010_003108
+2010_003112
+2010_003114
+2010_003115
+2010_003117
+2010_003119
+2010_003120
+2010_003122
+2010_003123
+2010_003127
+2010_003129
+2010_003132
+2010_003133
+2010_003135
+2010_003137
+2010_003138
+2010_003139
+2010_003143
+2010_003146
+2010_003147
+2010_003148
+2010_003149
+2010_003151
+2010_003153
+2010_003154
+2010_003156
+2010_003157
+2010_003159
+2010_003160
+2010_003162
+2010_003168
+2010_003169
+2010_003170
+2010_003173
+2010_003174
+2010_003176
+2010_003179
+2010_003183
+2010_003185
+2010_003186
+2010_003187
+2010_003190
+2010_003191
+2010_003192
+2010_003197
+2010_003199
+2010_003200
+2010_003201
+2010_003203
+2010_003204
+2010_003206
+2010_003207
+2010_003212
+2010_003214
+2010_003218
+2010_003219
+2010_003220
+2010_003222
+2010_003223
+2010_003227
+2010_003230
+2010_003231
+2010_003232
+2010_003233
+2010_003236
+2010_003238
+2010_003239
+2010_003240
+2010_003241
+2010_003244
+2010_003248
+2010_003249
+2010_003250
+2010_003251
+2010_003252
+2010_003253
+2010_003255
+2010_003256
+2010_003257
+2010_003259
+2010_003260
+2010_003263
+2010_003264
+2010_003269
+2010_003270
+2010_003274
+2010_003275
+2010_003276
+2010_003278
+2010_003279
+2010_003280
+2010_003283
+2010_003285
+2010_003287
+2010_003290
+2010_003291
+2010_003293
+2010_003297
+2010_003299
+2010_003300
+2010_003301
+2010_003302
+2010_003303
+2010_003304
+2010_003305
+2010_003309
+2010_003314
+2010_003316
+2010_003321
+2010_003325
+2010_003326
+2010_003329
+2010_003331
+2010_003332
+2010_003333
+2010_003335
+2010_003337
+2010_003341
+2010_003342
+2010_003343
+2010_003344
+2010_003345
+2010_003350
+2010_003351
+2010_003353
+2010_003355
+2010_003358
+2010_003361
+2010_003362
+2010_003365
+2010_003366
+2010_003367
+2010_003368
+2010_003370
+2010_003371
+2010_003372
+2010_003374
+2010_003375
+2010_003376
+2010_003379
+2010_003380
+2010_003381
+2010_003383
+2010_003384
+2010_003385
+2010_003390
+2010_003391
+2010_003395
+2010_003397
+2010_003398
+2010_003400
+2010_003401
+2010_003402
+2010_003405
+2010_003406
+2010_003409
+2010_003411
+2010_003415
+2010_003418
+2010_003419
+2010_003421
+2010_003427
+2010_003429
+2010_003432
+2010_003435
+2010_003436
+2010_003437
+2010_003439
+2010_003446
+2010_003450
+2010_003451
+2010_003453
+2010_003458
+2010_003461
+2010_003465
+2010_003467
+2010_003468
+2010_003469
+2010_003470
+2010_003473
+2010_003474
+2010_003477
+2010_003478
+2010_003479
+2010_003481
+2010_003482
+2010_003483
+2010_003488
+2010_003490
+2010_003491
+2010_003493
+2010_003495
+2010_003496
+2010_003497
+2010_003503
+2010_003506
+2010_003507
+2010_003508
+2010_003509
+2010_003512
+2010_003513
+2010_003514
+2010_003520
+2010_003522
+2010_003526
+2010_003527
+2010_003529
+2010_003531
+2010_003532
+2010_003534
+2010_003535
+2010_003537
+2010_003538
+2010_003539
+2010_003540
+2010_003541
+2010_003546
+2010_003547
+2010_003549
+2010_003551
+2010_003554
+2010_003556
+2010_003559
+2010_003560
+2010_003561
+2010_003562
+2010_003563
+2010_003567
+2010_003568
+2010_003569
+2010_003573
+2010_003574
+2010_003576
+2010_003579
+2010_003582
+2010_003585
+2010_003588
+2010_003592
+2010_003594
+2010_003597
+2010_003598
+2010_003599
+2010_003601
+2010_003603
+2010_003604
+2010_003605
+2010_003608
+2010_003609
+2010_003610
+2010_003612
+2010_003613
+2010_003618
+2010_003625
+2010_003628
+2010_003629
+2010_003630
+2010_003632
+2010_003634
+2010_003635
+2010_003640
+2010_003641
+2010_003643
+2010_003644
+2010_003645
+2010_003648
+2010_003649
+2010_003651
+2010_003653
+2010_003655
+2010_003656
+2010_003659
+2010_003664
+2010_003665
+2010_003667
+2010_003670
+2010_003671
+2010_003672
+2010_003673
+2010_003674
+2010_003675
+2010_003677
+2010_003679
+2010_003680
+2010_003686
+2010_003687
+2010_003688
+2010_003689
+2010_003690
+2010_003695
+2010_003696
+2010_003701
+2010_003703
+2010_003708
+2010_003709
+2010_003714
+2010_003716
+2010_003717
+2010_003719
+2010_003721
+2010_003723
+2010_003724
+2010_003725
+2010_003728
+2010_003729
+2010_003730
+2010_003731
+2010_003734
+2010_003735
+2010_003736
+2010_003737
+2010_003742
+2010_003743
+2010_003744
+2010_003745
+2010_003746
+2010_003747
+2010_003752
+2010_003754
+2010_003755
+2010_003757
+2010_003758
+2010_003761
+2010_003762
+2010_003764
+2010_003768
+2010_003770
+2010_003771
+2010_003772
+2010_003773
+2010_003774
+2010_003779
+2010_003781
+2010_003784
+2010_003788
+2010_003789
+2010_003791
+2010_003792
+2010_003798
+2010_003799
+2010_003800
+2010_003801
+2010_003804
+2010_003805
+2010_003806
+2010_003807
+2010_003811
+2010_003813
+2010_003815
+2010_003816
+2010_003818
+2010_003820
+2010_003821
+2010_003822
+2010_003823
+2010_003825
+2010_003826
+2010_003828
+2010_003837
+2010_003844
+2010_003845
+2010_003847
+2010_003848
+2010_003852
+2010_003854
+2010_003855
+2010_003856
+2010_003857
+2010_003859
+2010_003860
+2010_003861
+2010_003863
+2010_003864
+2010_003865
+2010_003871
+2010_003874
+2010_003875
+2010_003877
+2010_003878
+2010_003879
+2010_003884
+2010_003887
+2010_003890
+2010_003891
+2010_003892
+2010_003893
+2010_003894
+2010_003897
+2010_003898
+2010_003899
+2010_003900
+2010_003906
+2010_003910
+2010_003911
+2010_003912
+2010_003914
+2010_003915
+2010_003919
+2010_003920
+2010_003925
+2010_003928
+2010_003929
+2010_003931
+2010_003933
+2010_003936
+2010_003937
+2010_003938
+2010_003939
+2010_003942
+2010_003943
+2010_003944
+2010_003945
+2010_003947
+2010_003949
+2010_003950
+2010_003954
+2010_003955
+2010_003956
+2010_003957
+2010_003958
+2010_003961
+2010_003966
+2010_003970
+2010_003971
+2010_003974
+2010_003976
+2010_003980
+2010_003981
+2010_003982
+2010_003983
+2010_003987
+2010_003988
+2010_003994
+2010_003995
+2010_003996
+2010_003999
+2010_004002
+2010_004005
+2010_004006
+2010_004007
+2010_004008
+2010_004009
+2010_004010
+2010_004011
+2010_004014
+2010_004017
+2010_004021
+2010_004023
+2010_004025
+2010_004026
+2010_004027
+2010_004028
+2010_004029
+2010_004030
+2010_004031
+2010_004033
+2010_004036
+2010_004037
+2010_004041
+2010_004042
+2010_004043
+2010_004045
+2010_004048
+2010_004050
+2010_004052
+2010_004053
+2010_004054
+2010_004056
+2010_004059
+2010_004060
+2010_004061
+2010_004062
+2010_004063
+2010_004064
+2010_004065
+2010_004066
+2010_004067
+2010_004069
+2010_004071
+2010_004072
+2010_004073
+2010_004074
+2010_004075
+2010_004081
+2010_004084
+2010_004088
+2010_004089
+2010_004092
+2010_004094
+2010_004095
+2010_004096
+2010_004102
+2010_004104
+2010_004105
+2010_004107
+2010_004108
+2010_004109
+2010_004111
+2010_004116
+2010_004118
+2010_004119
+2010_004120
+2010_004121
+2010_004123
+2010_004124
+2010_004125
+2010_004129
+2010_004130
+2010_004133
+2010_004137
+2010_004138
+2010_004139
+2010_004140
+2010_004141
+2010_004143
+2010_004144
+2010_004145
+2010_004148
+2010_004149
+2010_004154
+2010_004157
+2010_004160
+2010_004161
+2010_004162
+2010_004163
+2010_004165
+2010_004168
+2010_004171
+2010_004172
+2010_004173
+2010_004175
+2010_004178
+2010_004179
+2010_004180
+2010_004182
+2010_004184
+2010_004186
+2010_004187
+2010_004188
+2010_004191
+2010_004192
+2010_004193
+2010_004197
+2010_004198
+2010_004201
+2010_004204
+2010_004207
+2010_004208
+2010_004209
+2010_004210
+2010_004211
+2010_004216
+2010_004219
+2010_004222
+2010_004223
+2010_004224
+2010_004225
+2010_004226
+2010_004227
+2010_004228
+2010_004229
+2010_004230
+2010_004231
+2010_004238
+2010_004239
+2010_004242
+2010_004244
+2010_004247
+2010_004248
+2010_004249
+2010_004252
+2010_004253
+2010_004254
+2010_004256
+2010_004257
+2010_004258
+2010_004259
+2010_004263
+2010_004264
+2010_004271
+2010_004275
+2010_004276
+2010_004278
+2010_004279
+2010_004280
+2010_004282
+2010_004283
+2010_004286
+2010_004288
+2010_004289
+2010_004290
+2010_004291
+2010_004295
+2010_004296
+2010_004297
+2010_004301
+2010_004304
+2010_004306
+2010_004307
+2010_004311
+2010_004312
+2010_004313
+2010_004314
+2010_004318
+2010_004320
+2010_004322
+2010_004325
+2010_004327
+2010_004332
+2010_004333
+2010_004335
+2010_004336
+2010_004337
+2010_004339
+2010_004341
+2010_004344
+2010_004345
+2010_004346
+2010_004348
+2010_004349
+2010_004350
+2010_004351
+2010_004352
+2010_004355
+2010_004357
+2010_004358
+2010_004360
+2010_004361
+2010_004362
+2010_004363
+2010_004365
+2010_004366
+2010_004367
+2010_004368
+2010_004369
+2010_004370
+2010_004371
+2010_004373
+2010_004374
+2010_004380
+2010_004382
+2010_004385
+2010_004387
+2010_004390
+2010_004391
+2010_004400
+2010_004402
+2010_004404
+2010_004409
+2010_004412
+2010_004415
+2010_004417
+2010_004419
+2010_004420
+2010_004422
+2010_004423
+2010_004425
+2010_004428
+2010_004429
+2010_004431
+2010_004432
+2010_004436
+2010_004439
+2010_004441
+2010_004445
+2010_004447
+2010_004448
+2010_004450
+2010_004451
+2010_004455
+2010_004456
+2010_004457
+2010_004459
+2010_004460
+2010_004461
+2010_004466
+2010_004467
+2010_004469
+2010_004472
+2010_004475
+2010_004476
+2010_004477
+2010_004478
+2010_004479
+2010_004481
+2010_004483
+2010_004484
+2010_004486
+2010_004488
+2010_004491
+2010_004492
+2010_004493
+2010_004499
+2010_004501
+2010_004503
+2010_004505
+2010_004506
+2010_004509
+2010_004511
+2010_004514
+2010_004515
+2010_004517
+2010_004518
+2010_004519
+2010_004520
+2010_004521
+2010_004523
+2010_004529
+2010_004533
+2010_004536
+2010_004537
+2010_004540
+2010_004542
+2010_004543
+2010_004545
+2010_004546
+2010_004550
+2010_004551
+2010_004553
+2010_004554
+2010_004556
+2010_004557
+2010_004558
+2010_004559
+2010_004560
+2010_004561
+2010_004567
+2010_004569
+2010_004570
+2010_004573
+2010_004575
+2010_004576
+2010_004577
+2010_004581
+2010_004584
+2010_004585
+2010_004586
+2010_004588
+2010_004591
+2010_004592
+2010_004594
+2010_004596
+2010_004597
+2010_004598
+2010_004600
+2010_004601
+2010_004604
+2010_004608
+2010_004609
+2010_004616
+2010_004618
+2010_004620
+2010_004621
+2010_004624
+2010_004625
+2010_004627
+2010_004628
+2010_004629
+2010_004631
+2010_004634
+2010_004635
+2010_004637
+2010_004638
+2010_004642
+2010_004646
+2010_004654
+2010_004655
+2010_004656
+2010_004657
+2010_004659
+2010_004660
+2010_004661
+2010_004662
+2010_004665
+2010_004666
+2010_004667
+2010_004669
+2010_004670
+2010_004672
+2010_004676
+2010_004677
+2010_004679
+2010_004680
+2010_004681
+2010_004683
+2010_004686
+2010_004690
+2010_004691
+2010_004692
+2010_004694
+2010_004696
+2010_004697
+2010_004698
+2010_004703
+2010_004704
+2010_004708
+2010_004710
+2010_004712
+2010_004714
+2010_004717
+2010_004721
+2010_004722
+2010_004726
+2010_004728
+2010_004729
+2010_004730
+2010_004733
+2010_004735
+2010_004738
+2010_004741
+2010_004743
+2010_004747
+2010_004748
+2010_004749
+2010_004750
+2010_004751
+2010_004753
+2010_004756
+2010_004757
+2010_004760
+2010_004763
+2010_004765
+2010_004766
+2010_004768
+2010_004770
+2010_004772
+2010_004773
+2010_004775
+2010_004777
+2010_004778
+2010_004779
+2010_004782
+2010_004783
+2010_004785
+2010_004786
+2010_004789
+2010_004791
+2010_004792
+2010_004793
+2010_004795
+2010_004797
+2010_004804
+2010_004805
+2010_004806
+2010_004807
+2010_004808
+2010_004809
+2010_004812
+2010_004813
+2010_004815
+2010_004816
+2010_004817
+2010_004821
+2010_004822
+2010_004824
+2010_004825
+2010_004826
+2010_004828
+2010_004829
+2010_004830
+2010_004831
+2010_004832
+2010_004836
+2010_004838
+2010_004841
+2010_004844
+2010_004847
+2010_004848
+2010_004849
+2010_004852
+2010_004854
+2010_004855
+2010_004856
+2010_004857
+2010_004861
+2010_004865
+2010_004866
+2010_004868
+2010_004871
+2010_004874
+2010_004877
+2010_004878
+2010_004879
+2010_004888
+2010_004889
+2010_004890
+2010_004891
+2010_004894
+2010_004896
+2010_004900
+2010_004901
+2010_004903
+2010_004906
+2010_004908
+2010_004909
+2010_004910
+2010_004913
+2010_004916
+2010_004917
+2010_004918
+2010_004919
+2010_004921
+2010_004922
+2010_004928
+2010_004930
+2010_004931
+2010_004933
+2010_004937
+2010_004938
+2010_004941
+2010_004942
+2010_004943
+2010_004944
+2010_004945
+2010_004946
+2010_004948
+2010_004950
+2010_004951
+2010_004952
+2010_004953
+2010_004954
+2010_004957
+2010_004959
+2010_004960
+2010_004962
+2010_004963
+2010_004966
+2010_004967
+2010_004968
+2010_004970
+2010_004971
+2010_004973
+2010_004974
+2010_004980
+2010_004982
+2010_004983
+2010_004987
+2010_004989
+2010_004991
+2010_004992
+2010_004994
+2010_004995
+2010_004997
+2010_004998
+2010_005000
+2010_005002
+2010_005005
+2010_005006
+2010_005008
+2010_005011
+2010_005013
+2010_005016
+2010_005017
+2010_005018
+2010_005019
+2010_005021
+2010_005022
+2010_005023
+2010_005026
+2010_005028
+2010_005031
+2010_005033
+2010_005035
+2010_005041
+2010_005042
+2010_005044
+2010_005046
+2010_005048
+2010_005049
+2010_005052
+2010_005053
+2010_005054
+2010_005055
+2010_005059
+2010_005060
+2010_005061
+2010_005062
+2010_005063
+2010_005064
+2010_005066
+2010_005068
+2010_005071
+2010_005072
+2010_005075
+2010_005079
+2010_005080
+2010_005082
+2010_005083
+2010_005087
+2010_005090
+2010_005093
+2010_005094
+2010_005096
+2010_005098
+2010_005099
+2010_005100
+2010_005101
+2010_005106
+2010_005107
+2010_005108
+2010_005109
+2010_005110
+2010_005111
+2010_005115
+2010_005116
+2010_005118
+2010_005119
+2010_005120
+2010_005123
+2010_005127
+2010_005128
+2010_005129
+2010_005130
+2010_005133
+2010_005134
+2010_005136
+2010_005138
+2010_005141
+2010_005143
+2010_005147
+2010_005148
+2010_005149
+2010_005152
+2010_005155
+2010_005158
+2010_005159
+2010_005160
+2010_005161
+2010_005164
+2010_005166
+2010_005167
+2010_005169
+2010_005170
+2010_005174
+2010_005180
+2010_005182
+2010_005183
+2010_005184
+2010_005185
+2010_005187
+2010_005188
+2010_005190
+2010_005192
+2010_005193
+2010_005198
+2010_005199
+2010_005201
+2010_005202
+2010_005206
+2010_005208
+2010_005211
+2010_005213
+2010_005215
+2010_005216
+2010_005217
+2010_005222
+2010_005223
+2010_005224
+2010_005226
+2010_005229
+2010_005230
+2010_005232
+2010_005236
+2010_005238
+2010_005239
+2010_005241
+2010_005242
+2010_005243
+2010_005245
+2010_005246
+2010_005250
+2010_005252
+2010_005253
+2010_005257
+2010_005258
+2010_005260
+2010_005261
+2010_005264
+2010_005266
+2010_005268
+2010_005270
+2010_005272
+2010_005273
+2010_005274
+2010_005275
+2010_005276
+2010_005277
+2010_005279
+2010_005284
+2010_005285
+2010_005287
+2010_005292
+2010_005293
+2010_005297
+2010_005299
+2010_005301
+2010_005303
+2010_005305
+2010_005306
+2010_005308
+2010_005309
+2010_005310
+2010_005312
+2010_005314
+2010_005317
+2010_005318
+2010_005320
+2010_005323
+2010_005327
+2010_005330
+2010_005331
+2010_005332
+2010_005338
+2010_005340
+2010_005344
+2010_005345
+2010_005346
+2010_005349
+2010_005350
+2010_005352
+2010_005353
+2010_005359
+2010_005361
+2010_005364
+2010_005365
+2010_005366
+2010_005369
+2010_005371
+2010_005372
+2010_005374
+2010_005375
+2010_005376
+2010_005377
+2010_005379
+2010_005382
+2010_005384
+2010_005385
+2010_005386
+2010_005388
+2010_005389
+2010_005391
+2010_005393
+2010_005394
+2010_005398
+2010_005401
+2010_005402
+2010_005403
+2010_005405
+2010_005406
+2010_005408
+2010_005409
+2010_005410
+2010_005414
+2010_005415
+2010_005416
+2010_005417
+2010_005419
+2010_005421
+2010_005424
+2010_005425
+2010_005426
+2010_005428
+2010_005429
+2010_005432
+2010_005433
+2010_005434
+2010_005437
+2010_005441
+2010_005442
+2010_005448
+2010_005450
+2010_005452
+2010_005455
+2010_005456
+2010_005457
+2010_005458
+2010_005462
+2010_005463
+2010_005466
+2010_005467
+2010_005468
+2010_005471
+2010_005472
+2010_005474
+2010_005475
+2010_005480
+2010_005482
+2010_005483
+2010_005484
+2010_005489
+2010_005491
+2010_005492
+2010_005493
+2010_005494
+2010_005496
+2010_005497
+2010_005498
+2010_005500
+2010_005501
+2010_005502
+2010_005505
+2010_005506
+2010_005508
+2010_005511
+2010_005512
+2010_005513
+2010_005514
+2010_005515
+2010_005516
+2010_005518
+2010_005519
+2010_005522
+2010_005527
+2010_005531
+2010_005532
+2010_005534
+2010_005535
+2010_005536
+2010_005538
+2010_005540
+2010_005542
+2010_005543
+2010_005546
+2010_005548
+2010_005551
+2010_005556
+2010_005557
+2010_005559
+2010_005561
+2010_005562
+2010_005565
+2010_005566
+2010_005567
+2010_005570
+2010_005571
+2010_005572
+2010_005573
+2010_005575
+2010_005576
+2010_005578
+2010_005582
+2010_005584
+2010_005585
+2010_005586
+2010_005587
+2010_005588
+2010_005591
+2010_005592
+2010_005593
+2010_005594
+2010_005595
+2010_005596
+2010_005597
+2010_005601
+2010_005603
+2010_005604
+2010_005606
+2010_005608
+2010_005610
+2010_005612
+2010_005614
+2010_005615
+2010_005616
+2010_005619
+2010_005620
+2010_005625
+2010_005626
+2010_005627
+2010_005628
+2010_005629
+2010_005632
+2010_005635
+2010_005636
+2010_005637
+2010_005640
+2010_005643
+2010_005644
+2010_005646
+2010_005647
+2010_005651
+2010_005652
+2010_005654
+2010_005657
+2010_005658
+2010_005663
+2010_005664
+2010_005665
+2010_005666
+2010_005668
+2010_005669
+2010_005670
+2010_005671
+2010_005672
+2010_005676
+2010_005678
+2010_005681
+2010_005683
+2010_005684
+2010_005688
+2010_005692
+2010_005696
+2010_005697
+2010_005700
+2010_005705
+2010_005706
+2010_005709
+2010_005712
+2010_005715
+2010_005716
+2010_005718
+2010_005719
+2010_005721
+2010_005723
+2010_005725
+2010_005727
+2010_005731
+2010_005732
+2010_005733
+2010_005734
+2010_005735
+2010_005736
+2010_005738
+2010_005740
+2010_005744
+2010_005746
+2010_005747
+2010_005748
+2010_005750
+2010_005752
+2010_005753
+2010_005754
+2010_005755
+2010_005756
+2010_005758
+2010_005761
+2010_005762
+2010_005763
+2010_005764
+2010_005767
+2010_005768
+2010_005770
+2010_005775
+2010_005776
+2010_005777
+2010_005780
+2010_005782
+2010_005784
+2010_005785
+2010_005788
+2010_005791
+2010_005794
+2010_005796
+2010_005800
+2010_005804
+2010_005805
+2010_005806
+2010_005807
+2010_005810
+2010_005815
+2010_005816
+2010_005817
+2010_005820
+2010_005821
+2010_005823
+2010_005824
+2010_005825
+2010_005826
+2010_005827
+2010_005830
+2010_005833
+2010_005835
+2010_005836
+2010_005837
+2010_005838
+2010_005840
+2010_005841
+2010_005843
+2010_005845
+2010_005847
+2010_005848
+2010_005849
+2010_005853
+2010_005854
+2010_005855
+2010_005860
+2010_005865
+2010_005867
+2010_005868
+2010_005870
+2010_005871
+2010_005874
+2010_005875
+2010_005876
+2010_005877
+2010_005882
+2010_005883
+2010_005884
+2010_005885
+2010_005886
+2010_005888
+2010_005891
+2010_005892
+2010_005894
+2010_005896
+2010_005897
+2010_005898
+2010_005899
+2010_005901
+2010_005903
+2010_005904
+2010_005906
+2010_005907
+2010_005909
+2010_005914
+2010_005919
+2010_005921
+2010_005922
+2010_005927
+2010_005928
+2010_005929
+2010_005930
+2010_005932
+2010_005934
+2010_005935
+2010_005936
+2010_005937
+2010_005938
+2010_005942
+2010_005943
+2010_005948
+2010_005949
+2010_005951
+2010_005952
+2010_005953
+2010_005954
+2010_005957
+2010_005958
+2010_005959
+2010_005960
+2010_005967
+2010_005968
+2010_005972
+2010_005973
+2010_005974
+2010_005975
+2010_005976
+2010_005978
+2010_005980
+2010_005981
+2010_005982
+2010_005984
+2010_005985
+2010_005986
+2010_005987
+2010_005991
+2010_005992
+2010_005993
+2010_005995
+2010_005996
+2010_005997
+2010_005998
+2010_006000
+2010_006003
+2010_006004
+2010_006009
+2010_006010
+2010_006011
+2010_006012
+2010_006015
+2010_006021
+2010_006023
+2010_006025
+2010_006026
+2010_006028
+2010_006031
+2010_006032
+2010_006033
+2010_006034
+2010_006035
+2010_006037
+2010_006040
+2010_006041
+2010_006042
+2010_006050
+2010_006051
+2010_006054
+2010_006056
+2010_006057
+2010_006058
+2010_006061
+2010_006062
+2010_006063
+2010_006066
+2010_006067
+2010_006070
+2010_006073
+2010_006076
+2010_006078
+2010_006079
+2010_006082
+2010_006084
+2010_006086
+2011_000002
+2011_000003
+2011_000006
+2011_000007
+2011_000009
+2011_000010
+2011_000012
+2011_000016
+2011_000017
+2011_000022
+2011_000025
+2011_000027
+2011_000028
+2011_000030
+2011_000034
+2011_000036
+2011_000037
+2011_000038
+2011_000041
+2011_000043
+2011_000044
+2011_000045
+2011_000048
+2011_000051
+2011_000052
+2011_000053
+2011_000054
+2011_000057
+2011_000058
+2011_000060
+2011_000061
+2011_000065
+2011_000066
+2011_000068
+2011_000069
+2011_000070
+2011_000071
+2011_000072
+2011_000076
+2011_000077
+2011_000082
+2011_000083
+2011_000084
+2011_000086
+2011_000087
+2011_000090
+2011_000094
+2011_000095
+2011_000096
+2011_000098
+2011_000102
+2011_000103
+2011_000105
+2011_000108
+2011_000109
+2011_000112
+2011_000114
+2011_000116
+2011_000122
+2011_000124
+2011_000128
+2011_000129
+2011_000130
+2011_000137
+2011_000138
+2011_000142
+2011_000145
+2011_000146
+2011_000147
+2011_000149
+2011_000152
+2011_000161
+2011_000162
+2011_000163
+2011_000165
+2011_000166
+2011_000173
+2011_000176
+2011_000178
+2011_000180
+2011_000181
+2011_000182
+2011_000185
+2011_000192
+2011_000194
+2011_000195
+2011_000196
+2011_000197
+2011_000202
+2011_000206
+2011_000208
+2011_000210
+2011_000213
+2011_000214
+2011_000216
+2011_000219
+2011_000220
+2011_000221
+2011_000222
+2011_000224
+2011_000226
+2011_000228
+2011_000229
+2011_000232
+2011_000233
+2011_000234
+2011_000238
+2011_000239
+2011_000241
+2011_000243
+2011_000246
+2011_000248
+2011_000249
+2011_000250
+2011_000252
+2011_000253
+2011_000257
+2011_000258
+2011_000267
+2011_000268
+2011_000269
+2011_000273
+2011_000276
+2011_000277
+2011_000278
+2011_000282
+2011_000283
+2011_000285
+2011_000286
+2011_000288
+2011_000290
+2011_000291
+2011_000293
+2011_000297
+2011_000299
+2011_000304
+2011_000305
+2011_000307
+2011_000309
+2011_000310
+2011_000312
+2011_000314
+2011_000315
+2011_000317
+2011_000319
+2011_000320
+2011_000321
+2011_000322
+2011_000324
+2011_000329
+2011_000332
+2011_000338
+2011_000342
+2011_000343
+2011_000344
+2011_000345
+2011_000346
+2011_000347
+2011_000359
+2011_000361
+2011_000362
+2011_000364
+2011_000369
+2011_000370
+2011_000374
+2011_000375
+2011_000376
+2011_000379
+2011_000382
+2011_000383
+2011_000385
+2011_000386
+2011_000388
+2011_000391
+2011_000392
+2011_000396
+2011_000397
+2011_000398
+2011_000399
+2011_000400
+2011_000404
+2011_000408
+2011_000412
+2011_000413
+2011_000416
+2011_000418
+2011_000419
+2011_000420
+2011_000426
+2011_000427
+2011_000428
+2011_000430
+2011_000432
+2011_000434
+2011_000435
+2011_000436
+2011_000438
+2011_000442
+2011_000444
+2011_000445
+2011_000449
+2011_000450
+2011_000453
+2011_000454
+2011_000455
+2011_000456
+2011_000457
+2011_000461
+2011_000465
+2011_000468
+2011_000469
+2011_000471
+2011_000472
+2011_000474
+2011_000475
+2011_000477
+2011_000479
+2011_000481
+2011_000482
+2011_000485
+2011_000487
+2011_000491
+2011_000492
+2011_000494
+2011_000496
+2011_000498
+2011_000499
+2011_000502
+2011_000503
+2011_000505
+2011_000509
+2011_000511
+2011_000512
+2011_000513
+2011_000514
+2011_000518
+2011_000519
+2011_000520
+2011_000521
+2011_000526
+2011_000530
+2011_000531
+2011_000532
+2011_000534
+2011_000536
+2011_000538
+2011_000541
+2011_000542
+2011_000548
+2011_000550
+2011_000551
+2011_000554
+2011_000556
+2011_000557
+2011_000558
+2011_000559
+2011_000560
+2011_000565
+2011_000566
+2011_000567
+2011_000569
+2011_000572
+2011_000573
+2011_000575
+2011_000577
+2011_000578
+2011_000579
+2011_000585
+2011_000586
+2011_000589
+2011_000592
+2011_000594
+2011_000596
+2011_000598
+2011_000600
+2011_000607
+2011_000608
+2011_000609
+2011_000612
+2011_000618
+2011_000621
+2011_000622
+2011_000627
+2011_000628
+2011_000629
+2011_000630
+2011_000631
+2011_000634
+2011_000637
+2011_000638
+2011_000641
+2011_000642
+2011_000646
+2011_000651
+2011_000652
+2011_000655
+2011_000656
+2011_000657
+2011_000658
+2011_000661
+2011_000666
+2011_000669
+2011_000673
+2011_000675
+2011_000679
+2011_000682
+2011_000683
+2011_000684
+2011_000685
+2011_000688
+2011_000689
+2011_000690
+2011_000692
+2011_000698
+2011_000701
+2011_000703
+2011_000704
+2011_000709
+2011_000711
+2011_000713
+2011_000718
+2011_000724
+2011_000725
+2011_000730
+2011_000731
+2011_000734
+2011_000743
+2011_000744
+2011_000745
+2011_000747
+2011_000748
+2011_000749
+2011_000753
+2011_000755
+2011_000757
+2011_000758
+2011_000759
+2011_000763
+2011_000765
+2011_000767
+2011_000768
+2011_000769
+2011_000770
+2011_000771
+2011_000772
+2011_000774
+2011_000778
+2011_000780
+2011_000784
+2011_000785
+2011_000788
+2011_000789
+2011_000790
+2011_000791
+2011_000793
+2011_000800
+2011_000804
+2011_000806
+2011_000807
+2011_000809
+2011_000813
+2011_000815
+2011_000819
+2011_000820
+2011_000823
+2011_000824
+2011_000827
+2011_000828
+2011_000829
+2011_000830
+2011_000831
+2011_000834
+2011_000837
+2011_000839
+2011_000840
+2011_000843
+2011_000845
+2011_000847
+2011_000848
+2011_000850
+2011_000851
+2011_000853
+2011_000855
+2011_000858
+2011_000859
+2011_000872
+2011_000874
+2011_000875
+2011_000882
+2011_000885
+2011_000887
+2011_000888
+2011_000893
+2011_000895
+2011_000897
+2011_000898
+2011_000899
+2011_000900
+2011_000901
+2011_000908
+2011_000909
+2011_000912
+2011_000917
+2011_000919
+2011_000920
+2011_000922
+2011_000927
+2011_000930
+2011_000932
+2011_000933
+2011_000934
+2011_000940
+2011_000944
+2011_000947
+2011_000950
+2011_000951
+2011_000953
+2011_000954
+2011_000957
+2011_000961
+2011_000965
+2011_000969
+2011_000971
+2011_000973
+2011_000975
+2011_000977
+2011_000979
+2011_000981
+2011_000982
+2011_000983
+2011_000986
+2011_000987
+2011_000990
+2011_000991
+2011_000996
+2011_000997
+2011_000999
+2011_001001
+2011_001004
+2011_001005
+2011_001008
+2011_001009
+2011_001010
+2011_001011
+2011_001014
+2011_001015
+2011_001016
+2011_001019
+2011_001020
+2011_001022
+2011_001023
+2011_001025
+2011_001027
+2011_001028
+2011_001029
+2011_001030
+2011_001031
+2011_001032
+2011_001033
+2011_001034
+2011_001036
+2011_001040
+2011_001044
+2011_001047
+2011_001052
+2011_001054
+2011_001055
+2011_001056
+2011_001058
+2011_001060
+2011_001062
+2011_001064
+2011_001066
+2011_001069
+2011_001071
+2011_001073
+2011_001079
+2011_001080
+2011_001081
+2011_001082
+2011_001084
+2011_001086
+2011_001091
+2011_001093
+2011_001097
+2011_001100
+2011_001105
+2011_001106
+2011_001107
+2011_001110
+2011_001111
+2011_001114
+2011_001116
+2011_001117
+2011_001123
+2011_001124
+2011_001126
+2011_001127
+2011_001128
+2011_001133
+2011_001134
+2011_001135
+2011_001136
+2011_001137
+2011_001138
+2011_001139
+2011_001144
+2011_001146
+2011_001149
+2011_001150
+2011_001152
+2011_001153
+2011_001158
+2011_001159
+2011_001160
+2011_001161
+2011_001163
+2011_001166
+2011_001167
+2011_001168
+2011_001169
+2011_001173
+2011_001175
+2011_001176
+2011_001188
+2011_001189
+2011_001190
+2011_001192
+2011_001193
+2011_001198
+2011_001201
+2011_001203
+2011_001208
+2011_001211
+2011_001213
+2011_001215
+2011_001216
+2011_001217
+2011_001220
+2011_001221
+2011_001223
+2011_001226
+2011_001227
+2011_001229
+2011_001232
+2011_001238
+2011_001240
+2011_001245
+2011_001246
+2011_001251
+2011_001252
+2011_001253
+2011_001254
+2011_001255
+2011_001257
+2011_001259
+2011_001260
+2011_001261
+2011_001263
+2011_001264
+2011_001266
+2011_001270
+2011_001271
+2011_001272
+2011_001276
+2011_001277
+2011_001281
+2011_001282
+2011_001283
+2011_001284
+2011_001285
+2011_001286
+2011_001287
+2011_001288
+2011_001290
+2011_001292
+2011_001295
+2011_001302
+2011_001304
+2011_001305
+2011_001310
+2011_001311
+2011_001313
+2011_001315
+2011_001318
+2011_001319
+2011_001320
+2011_001323
+2011_001326
+2011_001327
+2011_001329
+2011_001330
+2011_001333
+2011_001335
+2011_001336
+2011_001337
+2011_001341
+2011_001344
+2011_001346
+2011_001350
+2011_001354
+2011_001355
+2011_001357
+2011_001360
+2011_001366
+2011_001369
+2011_001370
+2011_001373
+2011_001375
+2011_001381
+2011_001382
+2011_001384
+2011_001387
+2011_001388
+2011_001389
+2011_001390
+2011_001394
+2011_001399
+2011_001400
+2011_001402
+2011_001404
+2011_001406
+2011_001407
+2011_001411
+2011_001412
+2011_001414
+2011_001416
+2011_001421
+2011_001422
+2011_001424
+2011_001432
+2011_001434
+2011_001440
+2011_001441
+2011_001447
+2011_001449
+2011_001451
+2011_001455
+2011_001456
+2011_001463
+2011_001464
+2011_001466
+2011_001467
+2011_001471
+2011_001475
+2011_001476
+2011_001479
+2011_001480
+2011_001489
+2011_001498
+2011_001501
+2011_001503
+2011_001505
+2011_001507
+2011_001508
+2011_001510
+2011_001514
+2011_001518
+2011_001519
+2011_001521
+2011_001524
+2011_001525
+2011_001526
+2011_001529
+2011_001530
+2011_001531
+2011_001532
+2011_001534
+2011_001535
+2011_001536
+2011_001537
+2011_001538
+2011_001541
+2011_001542
+2011_001543
+2011_001544
+2011_001546
+2011_001547
+2011_001549
+2011_001557
+2011_001558
+2011_001560
+2011_001566
+2011_001567
+2011_001568
+2011_001571
+2011_001572
+2011_001573
+2011_001582
+2011_001586
+2011_001589
+2011_001591
+2011_001592
+2011_001596
+2011_001597
+2011_001599
+2011_001600
+2011_001601
+2011_001602
+2011_001605
+2011_001606
+2011_001607
+2011_001608
+2011_001611
+2011_001612
+2011_001613
+2011_001614
+2011_001616
+2011_001618
+2011_001619
+2011_001620
+2011_001621
+2011_001622
+2011_001624
+2011_001625
+2011_001628
+2011_001629
+2011_001632
+2011_001641
+2011_001642
+2011_001643
+2011_001647
+2011_001649
+2011_001650
+2011_001652
+2011_001653
+2011_001655
+2011_001656
+2011_001662
+2011_001663
+2011_001665
+2011_001666
+2011_001669
+2011_001671
+2011_001673
+2011_001674
+2011_001678
+2011_001679
+2011_001689
+2011_001691
+2011_001693
+2011_001694
+2011_001695
+2011_001698
+2011_001699
+2011_001700
+2011_001705
+2011_001707
+2011_001708
+2011_001710
+2011_001712
+2011_001713
+2011_001714
+2011_001715
+2011_001716
+2011_001719
+2011_001720
+2011_001722
+2011_001726
+2011_001727
+2011_001730
+2011_001732
+2011_001733
+2011_001739
+2011_001740
+2011_001741
+2011_001745
+2011_001747
+2011_001748
+2011_001751
+2011_001753
+2011_001754
+2011_001755
+2011_001757
+2011_001764
+2011_001765
+2011_001766
+2011_001769
+2011_001770
+2011_001771
+2011_001775
+2011_001776
+2011_001779
+2011_001782
+2011_001785
+2011_001789
+2011_001790
+2011_001791
+2011_001793
+2011_001794
+2011_001796
+2011_001799
+2011_001800
+2011_001801
+2011_001805
+2011_001806
+2011_001810
+2011_001811
+2011_001812
+2011_001815
+2011_001819
+2011_001820
+2011_001822
+2011_001824
+2011_001825
+2011_001826
+2011_001827
+2011_001833
+2011_001834
+2011_001837
+2011_001840
+2011_001841
+2011_001842
+2011_001845
+2011_001847
+2011_001854
+2011_001855
+2011_001856
+2011_001858
+2011_001862
+2011_001863
+2011_001866
+2011_001868
+2011_001870
+2011_001871
+2011_001872
+2011_001873
+2011_001875
+2011_001876
+2011_001877
+2011_001880
+2011_001884
+2011_001885
+2011_001886
+2011_001889
+2011_001891
+2011_001893
+2011_001895
+2011_001896
+2011_001900
+2011_001901
+2011_001902
+2011_001904
+2011_001906
+2011_001910
+2011_001911
+2011_001914
+2011_001919
+2011_001920
+2011_001922
+2011_001924
+2011_001926
+2011_001927
+2011_001928
+2011_001929
+2011_001930
+2011_001932
+2011_001937
+2011_001938
+2011_001941
+2011_001942
+2011_001944
+2011_001945
+2011_001946
+2011_001949
+2011_001950
+2011_001951
+2011_001952
+2011_001956
+2011_001959
+2011_001961
+2011_001962
+2011_001964
+2011_001966
+2011_001967
+2011_001971
+2011_001972
+2011_001974
+2011_001975
+2011_001977
+2011_001980
+2011_001982
+2011_001984
+2011_001986
+2011_001987
+2011_001988
+2011_001989
+2011_001991
+2011_002002
+2011_002003
+2011_002004
+2011_002005
+2011_002006
+2011_002012
+2011_002016
+2011_002018
+2011_002019
+2011_002021
+2011_002022
+2011_002027
+2011_002031
+2011_002033
+2011_002034
+2011_002036
+2011_002038
+2011_002039
+2011_002040
+2011_002041
+2011_002042
+2011_002044
+2011_002045
+2011_002046
+2011_002047
+2011_002049
+2011_002050
+2011_002053
+2011_002055
+2011_002062
+2011_002063
+2011_002064
+2011_002073
+2011_002074
+2011_002075
+2011_002079
+2011_002085
+2011_002088
+2011_002091
+2011_002093
+2011_002096
+2011_002097
+2011_002098
+2011_002100
+2011_002102
+2011_002105
+2011_002106
+2011_002107
+2011_002108
+2011_002109
+2011_002110
+2011_002111
+2011_002113
+2011_002114
+2011_002116
+2011_002119
+2011_002121
+2011_002124
+2011_002128
+2011_002131
+2011_002132
+2011_002134
+2011_002135
+2011_002137
+2011_002142
+2011_002143
+2011_002144
+2011_002147
+2011_002148
+2011_002149
+2011_002150
+2011_002154
+2011_002156
+2011_002158
+2011_002159
+2011_002160
+2011_002163
+2011_002167
+2011_002169
+2011_002173
+2011_002174
+2011_002177
+2011_002178
+2011_002179
+2011_002184
+2011_002185
+2011_002186
+2011_002189
+2011_002192
+2011_002193
+2011_002200
+2011_002211
+2011_002215
+2011_002218
+2011_002221
+2011_002222
+2011_002223
+2011_002224
+2011_002227
+2011_002228
+2011_002230
+2011_002234
+2011_002236
+2011_002237
+2011_002239
+2011_002241
+2011_002244
+2011_002245
+2011_002246
+2011_002247
+2011_002248
+2011_002251
+2011_002252
+2011_002253
+2011_002260
+2011_002265
+2011_002268
+2011_002269
+2011_002270
+2011_002272
+2011_002273
+2011_002276
+2011_002278
+2011_002279
+2011_002280
+2011_002281
+2011_002284
+2011_002291
+2011_002292
+2011_002294
+2011_002295
+2011_002298
+2011_002300
+2011_002301
+2011_002303
+2011_002308
+2011_002312
+2011_002317
+2011_002318
+2011_002322
+2011_002324
+2011_002325
+2011_002327
+2011_002330
+2011_002335
+2011_002341
+2011_002343
+2011_002346
+2011_002347
+2011_002348
+2011_002350
+2011_002357
+2011_002358
+2011_002359
+2011_002362
+2011_002365
+2011_002366
+2011_002371
+2011_002379
+2011_002380
+2011_002381
+2011_002384
+2011_002385
+2011_002386
+2011_002387
+2011_002388
+2011_002389
+2011_002391
+2011_002393
+2011_002394
+2011_002395
+2011_002396
+2011_002397
+2011_002398
+2011_002402
+2011_002406
+2011_002407
+2011_002409
+2011_002410
+2011_002413
+2011_002414
+2011_002418
+2011_002419
+2011_002420
+2011_002421
+2011_002422
+2011_002429
+2011_002433
+2011_002435
+2011_002436
+2011_002443
+2011_002447
+2011_002448
+2011_002453
+2011_002455
+2011_002457
+2011_002458
+2011_002459
+2011_002460
+2011_002461
+2011_002462
+2011_002463
+2011_002464
+2011_002470
+2011_002474
+2011_002476
+2011_002479
+2011_002482
+2011_002484
+2011_002488
+2011_002490
+2011_002491
+2011_002492
+2011_002494
+2011_002495
+2011_002498
+2011_002503
+2011_002504
+2011_002505
+2011_002507
+2011_002509
+2011_002511
+2011_002514
+2011_002515
+2011_002516
+2011_002519
+2011_002520
+2011_002526
+2011_002528
+2011_002531
+2011_002532
+2011_002533
+2011_002535
+2011_002536
+2011_002542
+2011_002543
+2011_002548
+2011_002551
+2011_002552
+2011_002553
+2011_002554
+2011_002555
+2011_002556
+2011_002558
+2011_002559
+2011_002560
+2011_002561
+2011_002566
+2011_002567
+2011_002568
+2011_002571
+2011_002575
+2011_002578
+2011_002579
+2011_002582
+2011_002583
+2011_002584
+2011_002585
+2011_002588
+2011_002589
+2011_002590
+2011_002592
+2011_002594
+2011_002598
+2011_002601
+2011_002605
+2011_002606
+2011_002609
+2011_002610
+2011_002612
+2011_002614
+2011_002616
+2011_002617
+2011_002618
+2011_002620
+2011_002623
+2011_002624
+2011_002629
+2011_002631
+2011_002636
+2011_002638
+2011_002639
+2011_002640
+2011_002641
+2011_002644
+2011_002649
+2011_002650
+2011_002652
+2011_002656
+2011_002657
+2011_002658
+2011_002661
+2011_002662
+2011_002664
+2011_002673
+2011_002674
+2011_002675
+2011_002676
+2011_002677
+2011_002678
+2011_002685
+2011_002687
+2011_002694
+2011_002697
+2011_002699
+2011_002706
+2011_002709
+2011_002713
+2011_002714
+2011_002715
+2011_002717
+2011_002719
+2011_002724
+2011_002725
+2011_002726
+2011_002730
+2011_002738
+2011_002740
+2011_002742
+2011_002746
+2011_002748
+2011_002750
+2011_002751
+2011_002752
+2011_002754
+2011_002756
+2011_002760
+2011_002765
+2011_002766
+2011_002767
+2011_002770
+2011_002772
+2011_002775
+2011_002776
+2011_002779
+2011_002780
+2011_002782
+2011_002784
+2011_002786
+2011_002790
+2011_002795
+2011_002796
+2011_002798
+2011_002802
+2011_002803
+2011_002805
+2011_002808
+2011_002810
+2011_002811
+2011_002812
+2011_002814
+2011_002817
+2011_002818
+2011_002821
+2011_002823
+2011_002826
+2011_002830
+2011_002831
+2011_002833
+2011_002834
+2011_002838
+2011_002841
+2011_002842
+2011_002851
+2011_002852
+2011_002854
+2011_002863
+2011_002864
+2011_002867
+2011_002868
+2011_002870
+2011_002871
+2011_002872
+2011_002873
+2011_002879
+2011_002880
+2011_002881
+2011_002883
+2011_002884
+2011_002885
+2011_002887
+2011_002889
+2011_002890
+2011_002897
+2011_002900
+2011_002908
+2011_002911
+2011_002912
+2011_002913
+2011_002916
+2011_002917
+2011_002920
+2011_002921
+2011_002924
+2011_002925
+2011_002927
+2011_002929
+2011_002930
+2011_002932
+2011_002933
+2011_002935
+2011_002937
+2011_002938
+2011_002940
+2011_002942
+2011_002943
+2011_002944
+2011_002947
+2011_002949
+2011_002951
+2011_002953
+2011_002956
+2011_002958
+2011_002962
+2011_002965
+2011_002966
+2011_002967
+2011_002969
+2011_002970
+2011_002971
+2011_002974
+2011_002975
+2011_002978
+2011_002979
+2011_002983
+2011_002985
+2011_002987
+2011_002988
+2011_002992
+2011_002993
+2011_002994
+2011_002997
+2011_002999
+2011_003002
+2011_003003
+2011_003005
+2011_003010
+2011_003011
+2011_003012
+2011_003013
+2011_003016
+2011_003019
+2011_003020
+2011_003023
+2011_003025
+2011_003027
+2011_003028
+2011_003029
+2011_003030
+2011_003034
+2011_003038
+2011_003039
+2011_003041
+2011_003043
+2011_003044
+2011_003047
+2011_003048
+2011_003049
+2011_003050
+2011_003054
+2011_003055
+2011_003057
+2011_003059
+2011_003063
+2011_003065
+2011_003066
+2011_003073
+2011_003074
+2011_003076
+2011_003078
+2011_003079
+2011_003081
+2011_003085
+2011_003086
+2011_003089
+2011_003091
+2011_003097
+2011_003098
+2011_003103
+2011_003109
+2011_003111
+2011_003114
+2011_003115
+2011_003121
+2011_003124
+2011_003132
+2011_003134
+2011_003138
+2011_003141
+2011_003145
+2011_003146
+2011_003148
+2011_003149
+2011_003150
+2011_003151
+2011_003152
+2011_003154
+2011_003158
+2011_003159
+2011_003162
+2011_003163
+2011_003166
+2011_003167
+2011_003168
+2011_003169
+2011_003171
+2011_003176
+2011_003177
+2011_003182
+2011_003183
+2011_003184
+2011_003185
+2011_003187
+2011_003188
+2011_003192
+2011_003194
+2011_003197
+2011_003201
+2011_003205
+2011_003207
+2011_003211
+2011_003212
+2011_003213
+2011_003216
+2011_003220
+2011_003223
+2011_003228
+2011_003230
+2011_003232
+2011_003236
+2011_003238
+2011_003240
+2011_003242
+2011_003244
+2011_003246
+2011_003247
+2011_003253
+2011_003254
+2011_003255
+2011_003256
+2011_003259
+2011_003260
+2011_003261
+2011_003262
+2011_003269
+2011_003271
+2011_003274
+2011_003275
+2011_003276
diff --git a/ImageSets/Main/val.txt b/ImageSets/Main/val.txt
new file mode 100644
index 0000000..0349a31
--- /dev/null
+++ b/ImageSets/Main/val.txt
@@ -0,0 +1,8333 @@
+000005
+000007
+000009
+000016
+000019
+000020
+000021
+000024
+000030
+000039
+000041
+000046
+000050
+000051
+000052
+000060
+000063
+000065
+000072
+000081
+000093
+000095
+000099
+000101
+000102
+000107
+000109
+000110
+000113
+000117
+000118
+000120
+000121
+000123
+000125
+000130
+000131
+000132
+000142
+000143
+000146
+000150
+000156
+000158
+000165
+000169
+000170
+000177
+000180
+000184
+000190
+000203
+000208
+000210
+000211
+000214
+000215
+000218
+000221
+000224
+000229
+000232
+000233
+000236
+000241
+000244
+000245
+000246
+000249
+000251
+000257
+000266
+000268
+000269
+000270
+000275
+000285
+000289
+000298
+000302
+000303
+000304
+000305
+000308
+000318
+000321
+000322
+000323
+000328
+000329
+000332
+000336
+000338
+000340
+000343
+000352
+000354
+000363
+000373
+000374
+000380
+000381
+000396
+000403
+000408
+000417
+000419
+000420
+000424
+000427
+000428
+000433
+000435
+000439
+000443
+000448
+000459
+000460
+000461
+000462
+000464
+000480
+000482
+000483
+000486
+000491
+000492
+000494
+000498
+000499
+000500
+000501
+000509
+000513
+000514
+000515
+000520
+000523
+000530
+000531
+000540
+000543
+000545
+000563
+000564
+000579
+000581
+000582
+000588
+000591
+000598
+000599
+000601
+000608
+000610
+000613
+000619
+000626
+000628
+000637
+000645
+000647
+000653
+000656
+000660
+000661
+000663
+000667
+000675
+000676
+000677
+000682
+000684
+000686
+000690
+000694
+000702
+000705
+000707
+000712
+000713
+000714
+000717
+000720
+000728
+000730
+000738
+000742
+000746
+000748
+000750
+000752
+000755
+000756
+000760
+000763
+000771
+000772
+000776
+000777
+000780
+000782
+000786
+000787
+000791
+000794
+000797
+000799
+000800
+000802
+000806
+000808
+000814
+000815
+000816
+000826
+000831
+000832
+000834
+000842
+000843
+000847
+000848
+000854
+000855
+000857
+000862
+000863
+000868
+000872
+000874
+000876
+000878
+000879
+000880
+000882
+000885
+000895
+000896
+000903
+000911
+000917
+000918
+000920
+000921
+000923
+000926
+000931
+000934
+000935
+000937
+000946
+000947
+000948
+000949
+000971
+000972
+000973
+000982
+001004
+001009
+001012
+001017
+001018
+001027
+001028
+001041
+001042
+001045
+001052
+001053
+001056
+001061
+001062
+001066
+001069
+001072
+001074
+001083
+001084
+001091
+001092
+001093
+001097
+001102
+001104
+001107
+001109
+001110
+001121
+001124
+001125
+001136
+001137
+001142
+001143
+001144
+001145
+001148
+001149
+001154
+001160
+001161
+001164
+001166
+001170
+001175
+001176
+001184
+001185
+001186
+001187
+001192
+001199
+001200
+001201
+001203
+001206
+001211
+001215
+001221
+001224
+001225
+001231
+001233
+001236
+001241
+001247
+001250
+001254
+001259
+001260
+001265
+001266
+001272
+001274
+001277
+001281
+001284
+001286
+001288
+001289
+001290
+001292
+001293
+001298
+001310
+001311
+001316
+001324
+001330
+001337
+001341
+001343
+001350
+001352
+001360
+001361
+001362
+001371
+001375
+001383
+001386
+001387
+001397
+001400
+001413
+001430
+001432
+001439
+001441
+001443
+001444
+001445
+001460
+001463
+001464
+001465
+001466
+001467
+001472
+001475
+001481
+001484
+001490
+001493
+001497
+001509
+001510
+001514
+001522
+001523
+001531
+001536
+001537
+001541
+001543
+001544
+001545
+001553
+001554
+001561
+001565
+001571
+001577
+001582
+001588
+001595
+001598
+001603
+001608
+001614
+001617
+001618
+001628
+001632
+001638
+001640
+001642
+001647
+001653
+001675
+001677
+001678
+001682
+001685
+001686
+001689
+001691
+001693
+001718
+001724
+001725
+001726
+001727
+001730
+001746
+001747
+001749
+001755
+001756
+001771
+001772
+001775
+001778
+001782
+001784
+001785
+001793
+001795
+001797
+001799
+001801
+001807
+001816
+001818
+001827
+001830
+001833
+001837
+001842
+001847
+001849
+001855
+001860
+001862
+001872
+001875
+001877
+001878
+001882
+001887
+001888
+001899
+001901
+001907
+001911
+001918
+001920
+001927
+001931
+001932
+001933
+001934
+001936
+001940
+001944
+001948
+001958
+001962
+001964
+001970
+001972
+001976
+001982
+002000
+002011
+002019
+002021
+002022
+002023
+002024
+002030
+002036
+002045
+002054
+002058
+002063
+002064
+002067
+002070
+002082
+002083
+002086
+002088
+002090
+002091
+002094
+002098
+002099
+002101
+002102
+002109
+002112
+002114
+002124
+002125
+002129
+002135
+002136
+002140
+002142
+002145
+002146
+002152
+002163
+002165
+002169
+002171
+002174
+002181
+002183
+002184
+002190
+002201
+002202
+002209
+002213
+002214
+002218
+002220
+002226
+002228
+002233
+002244
+002248
+002251
+002257
+002259
+002261
+002263
+002266
+002267
+002268
+002270
+002272
+002273
+002276
+002278
+002281
+002285
+002288
+002290
+002300
+002302
+002305
+002308
+002324
+002328
+002329
+002330
+002332
+002333
+002337
+002340
+002343
+002345
+002348
+002352
+002361
+002364
+002366
+002367
+002369
+002371
+002372
+002374
+002375
+002376
+002377
+002378
+002382
+002385
+002387
+002391
+002393
+002404
+002407
+002415
+002417
+002425
+002427
+002435
+002437
+002441
+002444
+002450
+002452
+002454
+002456
+002459
+002460
+002462
+002470
+002476
+002477
+002479
+002491
+002492
+002493
+002497
+002504
+002505
+002508
+002513
+002520
+002523
+002524
+002525
+002529
+002537
+002540
+002542
+002546
+002549
+002561
+002563
+002565
+002566
+002567
+002578
+002584
+002585
+002586
+002589
+002593
+002598
+002600
+002605
+002606
+002613
+002615
+002618
+002621
+002632
+002633
+002636
+002637
+002641
+002643
+002646
+002649
+002657
+002658
+002659
+002667
+002668
+002670
+002675
+002677
+002678
+002689
+002690
+002693
+002695
+002696
+002699
+002706
+002709
+002714
+002717
+002718
+002721
+002723
+002727
+002732
+002734
+002741
+002747
+002751
+002760
+002762
+002767
+002772
+002775
+002776
+002784
+002785
+002786
+002794
+002798
+002800
+002803
+002810
+002812
+002815
+002827
+002833
+002835
+002836
+002838
+002842
+002847
+002854
+002859
+002875
+002879
+002880
+002884
+002886
+002889
+002891
+002893
+002896
+002901
+002910
+002912
+002913
+002915
+002916
+002917
+002924
+002932
+002933
+002935
+002938
+002940
+002941
+002942
+002943
+002944
+002946
+002947
+002952
+002954
+002960
+002963
+002965
+002966
+002967
+002977
+002978
+002984
+002986
+002994
+003000
+003004
+003005
+003008
+003009
+003015
+003017
+003021
+003023
+003028
+003031
+003032
+003038
+003039
+003044
+003045
+003054
+003056
+003057
+003058
+003064
+003065
+003072
+003078
+003082
+003086
+003089
+003090
+003093
+003094
+003098
+003102
+003112
+003117
+003118
+003120
+003121
+003126
+003127
+003129
+003137
+003142
+003154
+003162
+003164
+003170
+003176
+003177
+003178
+003186
+003189
+003194
+003195
+003199
+003200
+003207
+003210
+003213
+003216
+003218
+003219
+003223
+003228
+003239
+003243
+003250
+003255
+003256
+003258
+003262
+003271
+003272
+003274
+003285
+003293
+003294
+003296
+003299
+003300
+003301
+003307
+003311
+003313
+003316
+003325
+003327
+003335
+003344
+003351
+003360
+003362
+003370
+003376
+003377
+003386
+003390
+003391
+003397
+003398
+003403
+003404
+003407
+003410
+003415
+003419
+003422
+003425
+003429
+003435
+003443
+003444
+003449
+003451
+003453
+003455
+003458
+003461
+003462
+003464
+003465
+003468
+003469
+003470
+003492
+003516
+003518
+003519
+003521
+003528
+003530
+003536
+003537
+003546
+003554
+003556
+003566
+003567
+003580
+003587
+003589
+003593
+003594
+003597
+003606
+003611
+003618
+003620
+003623
+003632
+003636
+003638
+003639
+003640
+003648
+003651
+003654
+003655
+003657
+003660
+003667
+003669
+003673
+003674
+003675
+003684
+003685
+003690
+003691
+003696
+003703
+003706
+003708
+003709
+003711
+003717
+003721
+003722
+003727
+003729
+003750
+003753
+003754
+003760
+003772
+003774
+003780
+003783
+003791
+003793
+003796
+003798
+003803
+003808
+003809
+003814
+003820
+003821
+003826
+003837
+003838
+003844
+003845
+003846
+003848
+003855
+003857
+003863
+003868
+003869
+003871
+003872
+003876
+003877
+003885
+003886
+003891
+003895
+003905
+003911
+003915
+003918
+003919
+003923
+003924
+003926
+003937
+003941
+003946
+003947
+003948
+003954
+003957
+003960
+003963
+003965
+003966
+003973
+003979
+003984
+003986
+003990
+003992
+003994
+003996
+004003
+004010
+004011
+004015
+004020
+004025
+004031
+004039
+004047
+004051
+004057
+004060
+004066
+004069
+004073
+004075
+004076
+004077
+004082
+004085
+004087
+004089
+004102
+004105
+004108
+004110
+004113
+004117
+004122
+004135
+004141
+004142
+004143
+004145
+004148
+004150
+004174
+004178
+004185
+004186
+004191
+004192
+004193
+004194
+004195
+004203
+004204
+004205
+004212
+004229
+004230
+004239
+004246
+004257
+004258
+004259
+004264
+004265
+004274
+004275
+004279
+004284
+004286
+004293
+004295
+004298
+004304
+004310
+004312
+004321
+004323
+004326
+004329
+004331
+004341
+004346
+004349
+004351
+004352
+004354
+004356
+004364
+004368
+004369
+004380
+004384
+004390
+004396
+004397
+004405
+004409
+004411
+004421
+004423
+004424
+004429
+004430
+004432
+004433
+004437
+004438
+004446
+004450
+004455
+004457
+004459
+004463
+004464
+004466
+004468
+004474
+004487
+004488
+004490
+004493
+004494
+004495
+004498
+004499
+004507
+004509
+004512
+004518
+004527
+004528
+004530
+004532
+004535
+004539
+004542
+004552
+004555
+004558
+004574
+004581
+004585
+004588
+004592
+004600
+004601
+004606
+004609
+004618
+004626
+004630
+004632
+004647
+004649
+004652
+004653
+004654
+004655
+004660
+004662
+004672
+004673
+004674
+004676
+004682
+004689
+004692
+004699
+004707
+004708
+004719
+004722
+004727
+004732
+004746
+004750
+004761
+004768
+004770
+004777
+004785
+004786
+004788
+004789
+004796
+004805
+004812
+004814
+004816
+004818
+004825
+004826
+004831
+004834
+004839
+004840
+004850
+004852
+004856
+004859
+004863
+004866
+004867
+004868
+004872
+004878
+004886
+004890
+004895
+004896
+004903
+004912
+004916
+004926
+004928
+004931
+004935
+004936
+004938
+004939
+004943
+004948
+004950
+004953
+004954
+004956
+004960
+004963
+004967
+004977
+004982
+004983
+004985
+004986
+004994
+004997
+004998
+004999
+005003
+005014
+005028
+005036
+005037
+005039
+005042
+005054
+005055
+005056
+005062
+005063
+005064
+005067
+005072
+005077
+005079
+005081
+005085
+005102
+005104
+005110
+005111
+005116
+005128
+005131
+005135
+005136
+005144
+005145
+005146
+005150
+005159
+005160
+005161
+005175
+005176
+005179
+005185
+005195
+005199
+005209
+005210
+005212
+005214
+005220
+005222
+005224
+005229
+005230
+005239
+005242
+005248
+005253
+005254
+005263
+005264
+005267
+005268
+005270
+005274
+005278
+005281
+005293
+005298
+005305
+005306
+005312
+005314
+005315
+005319
+005320
+005325
+005326
+005328
+005331
+005340
+005343
+005346
+005348
+005349
+005350
+005352
+005355
+005365
+005367
+005370
+005371
+005378
+005379
+005380
+005383
+005384
+005385
+005393
+005395
+005397
+005398
+005407
+005416
+005418
+005419
+005421
+005423
+005429
+005430
+005431
+005434
+005436
+005438
+005439
+005441
+005454
+005461
+005465
+005469
+005470
+005471
+005475
+005481
+005485
+005486
+005497
+005507
+005510
+005517
+005518
+005521
+005522
+005530
+005531
+005535
+005539
+005549
+005550
+005552
+005554
+005559
+005573
+005576
+005577
+005583
+005584
+005586
+005588
+005590
+005593
+005606
+005608
+005613
+005614
+005615
+005618
+005620
+005629
+005640
+005641
+005645
+005647
+005652
+005653
+005655
+005657
+005660
+005662
+005664
+005672
+005674
+005676
+005679
+005682
+005685
+005687
+005693
+005696
+005701
+005702
+005714
+005716
+005719
+005723
+005729
+005732
+005736
+005741
+005743
+005747
+005749
+005755
+005760
+005761
+005762
+005768
+005773
+005779
+005781
+005788
+005790
+005791
+005794
+005799
+005811
+005812
+005815
+005818
+005819
+005825
+005828
+005829
+005830
+005839
+005841
+005845
+005852
+005853
+005854
+005856
+005863
+005868
+005874
+005875
+005877
+005878
+005879
+005894
+005897
+005906
+005912
+005914
+005917
+005919
+005928
+005940
+005952
+005954
+005956
+005963
+005968
+005970
+005975
+005979
+005981
+005985
+005988
+005989
+005991
+005995
+005996
+005998
+006000
+006001
+006005
+006012
+006018
+006026
+006027
+006028
+006029
+006035
+006041
+006042
+006045
+006046
+006055
+006058
+006062
+006069
+006071
+006084
+006089
+006097
+006098
+006107
+006108
+006111
+006117
+006120
+006124
+006125
+006129
+006133
+006136
+006139
+006146
+006148
+006150
+006151
+006153
+006159
+006161
+006163
+006184
+006185
+006188
+006190
+006198
+006201
+006202
+006203
+006206
+006209
+006212
+006214
+006215
+006216
+006218
+006219
+006220
+006222
+006233
+006234
+006235
+006240
+006241
+006249
+006252
+006254
+006258
+006259
+006260
+006269
+006276
+006277
+006281
+006282
+006284
+006286
+006295
+006296
+006300
+006301
+006306
+006309
+006314
+006318
+006319
+006321
+006323
+006325
+006330
+006335
+006337
+006338
+006339
+006346
+006348
+006350
+006351
+006355
+006357
+006377
+006385
+006387
+006391
+006392
+006396
+006398
+006404
+006409
+006421
+006424
+006425
+006428
+006430
+006437
+006440
+006443
+006444
+006445
+006449
+006450
+006456
+006463
+006465
+006468
+006473
+006480
+006484
+006488
+006492
+006497
+006507
+006509
+006512
+006519
+006520
+006529
+006530
+006532
+006534
+006538
+006542
+006543
+006553
+006562
+006565
+006570
+006572
+006575
+006576
+006578
+006583
+006584
+006585
+006587
+006588
+006593
+006599
+006603
+006606
+006611
+006617
+006618
+006619
+006621
+006625
+006628
+006631
+006632
+006643
+006645
+006647
+006657
+006661
+006664
+006666
+006667
+006668
+006670
+006671
+006673
+006677
+006678
+006679
+006681
+006682
+006687
+006690
+006696
+006699
+006702
+006709
+006718
+006719
+006722
+006725
+006730
+006739
+006747
+006751
+006759
+006760
+006761
+006762
+006765
+006768
+006769
+006772
+006783
+006786
+006789
+006797
+006799
+006800
+006802
+006803
+006808
+006813
+006814
+006819
+006821
+006827
+006828
+006829
+006835
+006838
+006841
+006842
+006850
+006855
+006859
+006860
+006862
+006865
+006867
+006876
+006878
+006880
+006884
+006886
+006892
+006903
+006908
+006918
+006922
+006924
+006932
+006933
+006934
+006935
+006940
+006944
+006945
+006949
+006952
+006953
+006956
+006962
+006963
+006965
+006966
+006972
+006981
+006987
+006988
+006989
+006990
+006994
+006995
+007004
+007008
+007009
+007020
+007021
+007022
+007031
+007035
+007038
+007042
+007046
+007048
+007049
+007052
+007054
+007056
+007058
+007059
+007065
+007068
+007070
+007071
+007074
+007077
+007084
+007086
+007097
+007100
+007101
+007104
+007109
+007114
+007117
+007122
+007123
+007132
+007139
+007140
+007141
+007144
+007146
+007147
+007148
+007149
+007153
+007162
+007165
+007167
+007172
+007174
+007187
+007189
+007191
+007200
+007204
+007208
+007210
+007211
+007212
+007215
+007216
+007217
+007224
+007227
+007230
+007236
+007244
+007245
+007247
+007249
+007258
+007259
+007260
+007266
+007270
+007274
+007275
+007276
+007280
+007283
+007284
+007292
+007294
+007296
+007297
+007299
+007300
+007302
+007311
+007314
+007318
+007329
+007330
+007343
+007344
+007346
+007350
+007356
+007359
+007363
+007372
+007374
+007376
+007383
+007388
+007390
+007408
+007414
+007416
+007422
+007424
+007427
+007432
+007433
+007435
+007436
+007438
+007439
+007443
+007445
+007448
+007449
+007451
+007457
+007460
+007461
+007465
+007470
+007475
+007480
+007482
+007484
+007486
+007489
+007498
+007506
+007511
+007517
+007523
+007525
+007527
+007528
+007533
+007537
+007543
+007546
+007547
+007551
+007555
+007559
+007563
+007568
+007571
+007576
+007579
+007585
+007592
+007603
+007605
+007612
+007614
+007615
+007618
+007622
+007624
+007626
+007639
+007640
+007642
+007647
+007649
+007650
+007656
+007657
+007662
+007664
+007666
+007668
+007670
+007671
+007672
+007673
+007675
+007677
+007678
+007679
+007680
+007682
+007687
+007688
+007691
+007694
+007702
+007705
+007709
+007712
+007715
+007720
+007723
+007724
+007727
+007732
+007742
+007743
+007745
+007746
+007754
+007758
+007760
+007763
+007765
+007768
+007772
+007773
+007776
+007779
+007786
+007793
+007798
+007799
+007812
+007813
+007815
+007824
+007826
+007833
+007834
+007841
+007843
+007845
+007855
+007856
+007857
+007865
+007868
+007869
+007873
+007886
+007889
+007890
+007897
+007899
+007902
+007909
+007916
+007919
+007920
+007921
+007924
+007928
+007931
+007933
+007935
+007943
+007946
+007947
+007950
+007954
+007956
+007958
+007970
+007971
+007979
+007984
+007987
+007997
+007998
+007999
+008002
+008009
+008023
+008024
+008029
+008031
+008032
+008033
+008036
+008048
+008057
+008060
+008061
+008068
+008069
+008085
+008086
+008087
+008091
+008100
+008101
+008103
+008105
+008107
+008112
+008115
+008122
+008125
+008132
+008138
+008140
+008141
+008144
+008151
+008159
+008160
+008168
+008171
+008173
+008175
+008177
+008180
+008189
+008190
+008191
+008200
+008208
+008209
+008220
+008222
+008224
+008225
+008229
+008236
+008241
+008244
+008251
+008258
+008268
+008275
+008279
+008281
+008284
+008285
+008292
+008293
+008294
+008295
+008297
+008299
+008300
+008306
+008307
+008318
+008319
+008320
+008323
+008326
+008327
+008329
+008335
+008345
+008349
+008355
+008359
+008364
+008365
+008368
+008370
+008376
+008386
+008387
+008390
+008410
+008413
+008415
+008416
+008423
+008424
+008429
+008430
+008433
+008434
+008438
+008444
+008450
+008454
+008461
+008472
+008484
+008485
+008492
+008494
+008498
+008499
+008502
+008503
+008509
+008512
+008513
+008514
+008518
+008519
+008521
+008522
+008524
+008526
+008534
+008535
+008541
+008542
+008553
+008556
+008557
+008562
+008564
+008572
+008573
+008576
+008582
+008584
+008586
+008592
+008601
+008604
+008606
+008607
+008608
+008612
+008620
+008621
+008624
+008635
+008636
+008638
+008639
+008644
+008647
+008653
+008654
+008667
+008680
+008683
+008687
+008692
+008695
+008698
+008701
+008709
+008713
+008716
+008717
+008718
+008722
+008728
+008730
+008733
+008739
+008742
+008747
+008749
+008752
+008753
+008759
+008766
+008769
+008772
+008773
+008775
+008793
+008796
+008799
+008801
+008805
+008810
+008817
+008822
+008823
+008826
+008831
+008833
+008835
+008836
+008837
+008843
+008848
+008849
+008854
+008858
+008859
+008867
+008871
+008873
+008874
+008876
+008880
+008884
+008888
+008890
+008892
+008911
+008913
+008914
+008917
+008919
+008921
+008927
+008931
+008940
+008942
+008943
+008951
+008953
+008955
+008965
+008976
+008982
+008983
+008997
+009002
+009006
+009007
+009015
+009019
+009022
+009024
+009034
+009035
+009037
+009039
+009048
+009051
+009053
+009060
+009064
+009072
+009079
+009085
+009087
+009089
+009091
+009094
+009105
+009112
+009113
+009116
+009126
+009128
+009129
+009131
+009133
+009138
+009141
+009147
+009150
+009151
+009155
+009157
+009159
+009162
+009163
+009168
+009174
+009177
+009178
+009179
+009180
+009186
+009187
+009189
+009192
+009193
+009194
+009195
+009202
+009212
+009213
+009221
+009224
+009236
+009239
+009244
+009246
+009247
+009249
+009250
+009254
+009268
+009273
+009278
+009279
+009281
+009282
+009286
+009291
+009303
+009309
+009312
+009315
+009323
+009326
+009330
+009331
+009334
+009337
+009347
+009348
+009349
+009350
+009351
+009354
+009368
+009371
+009373
+009374
+009375
+009378
+009382
+009401
+009405
+009408
+009412
+009414
+009433
+009437
+009438
+009439
+009440
+009443
+009445
+009448
+009454
+009455
+009456
+009457
+009459
+009461
+009464
+009468
+009470
+009472
+009477
+009479
+009480
+009481
+009484
+009494
+009500
+009502
+009507
+009517
+009519
+009527
+009531
+009532
+009533
+009540
+009543
+009546
+009550
+009558
+009560
+009565
+009567
+009568
+009571
+009580
+009586
+009588
+009591
+009597
+009598
+009603
+009611
+009617
+009619
+009620
+009627
+009636
+009641
+009647
+009649
+009655
+009658
+009667
+009670
+009676
+009678
+009681
+009685
+009686
+009687
+009692
+009695
+009698
+009699
+009700
+009706
+009710
+009711
+009712
+009719
+009724
+009726
+009732
+009737
+009738
+009743
+009745
+009746
+009747
+009748
+009754
+009758
+009761
+009764
+009767
+009772
+009773
+009778
+009780
+009781
+009785
+009794
+009796
+009801
+009809
+009816
+009819
+009822
+009823
+009831
+009833
+009836
+009841
+009858
+009862
+009863
+009865
+009870
+009880
+009881
+009886
+009894
+009897
+009898
+009900
+009902
+009905
+009908
+009913
+009917
+009923
+009932
+009935
+009939
+009946
+009947
+009950
+009954
+009955
+009958
+2008_000002
+2008_000003
+2008_000007
+2008_000009
+2008_000016
+2008_000021
+2008_000026
+2008_000027
+2008_000032
+2008_000034
+2008_000042
+2008_000043
+2008_000050
+2008_000051
+2008_000052
+2008_000054
+2008_000056
+2008_000059
+2008_000062
+2008_000064
+2008_000067
+2008_000073
+2008_000075
+2008_000076
+2008_000078
+2008_000080
+2008_000082
+2008_000084
+2008_000090
+2008_000107
+2008_000115
+2008_000116
+2008_000119
+2008_000120
+2008_000123
+2008_000133
+2008_000134
+2008_000138
+2008_000140
+2008_000145
+2008_000149
+2008_000163
+2008_000174
+2008_000177
+2008_000182
+2008_000183
+2008_000190
+2008_000194
+2008_000195
+2008_000203
+2008_000204
+2008_000213
+2008_000215
+2008_000219
+2008_000222
+2008_000223
+2008_000233
+2008_000234
+2008_000239
+2008_000243
+2008_000244
+2008_000246
+2008_000251
+2008_000253
+2008_000254
+2008_000257
+2008_000261
+2008_000264
+2008_000268
+2008_000270
+2008_000271
+2008_000272
+2008_000274
+2008_000277
+2008_000278
+2008_000281
+2008_000298
+2008_000304
+2008_000305
+2008_000306
+2008_000307
+2008_000321
+2008_000328
+2008_000339
+2008_000340
+2008_000345
+2008_000354
+2008_000358
+2008_000359
+2008_000367
+2008_000373
+2008_000376
+2008_000378
+2008_000381
+2008_000382
+2008_000383
+2008_000391
+2008_000398
+2008_000401
+2008_000403
+2008_000406
+2008_000407
+2008_000408
+2008_000413
+2008_000414
+2008_000418
+2008_000419
+2008_000423
+2008_000424
+2008_000446
+2008_000452
+2008_000457
+2008_000464
+2008_000465
+2008_000466
+2008_000469
+2008_000472
+2008_000473
+2008_000474
+2008_000475
+2008_000481
+2008_000489
+2008_000492
+2008_000496
+2008_000498
+2008_000501
+2008_000510
+2008_000511
+2008_000516
+2008_000519
+2008_000522
+2008_000532
+2008_000533
+2008_000535
+2008_000536
+2008_000541
+2008_000547
+2008_000553
+2008_000558
+2008_000562
+2008_000564
+2008_000566
+2008_000568
+2008_000569
+2008_000573
+2008_000579
+2008_000581
+2008_000589
+2008_000599
+2008_000602
+2008_000605
+2008_000609
+2008_000614
+2008_000620
+2008_000622
+2008_000623
+2008_000629
+2008_000630
+2008_000634
+2008_000640
+2008_000647
+2008_000652
+2008_000656
+2008_000657
+2008_000659
+2008_000660
+2008_000661
+2008_000662
+2008_000666
+2008_000670
+2008_000673
+2008_000677
+2008_000690
+2008_000691
+2008_000695
+2008_000697
+2008_000699
+2008_000700
+2008_000705
+2008_000706
+2008_000714
+2008_000725
+2008_000727
+2008_000731
+2008_000734
+2008_000737
+2008_000740
+2008_000745
+2008_000748
+2008_000765
+2008_000769
+2008_000776
+2008_000780
+2008_000782
+2008_000783
+2008_000788
+2008_000793
+2008_000795
+2008_000796
+2008_000803
+2008_000804
+2008_000805
+2008_000806
+2008_000811
+2008_000817
+2008_000825
+2008_000828
+2008_000834
+2008_000835
+2008_000837
+2008_000839
+2008_000848
+2008_000853
+2008_000857
+2008_000858
+2008_000863
+2008_000864
+2008_000868
+2008_000876
+2008_000878
+2008_000880
+2008_000884
+2008_000885
+2008_000897
+2008_000904
+2008_000910
+2008_000911
+2008_000916
+2008_000917
+2008_000919
+2008_000922
+2008_000931
+2008_000936
+2008_000939
+2008_000940
+2008_000942
+2008_000943
+2008_000950
+2008_000952
+2008_000956
+2008_000957
+2008_000960
+2008_000964
+2008_000965
+2008_000971
+2008_000972
+2008_000976
+2008_000982
+2008_000984
+2008_000992
+2008_000993
+2008_001004
+2008_001007
+2008_001009
+2008_001012
+2008_001013
+2008_001024
+2008_001028
+2008_001034
+2008_001040
+2008_001041
+2008_001046
+2008_001055
+2008_001060
+2008_001062
+2008_001063
+2008_001066
+2008_001068
+2008_001070
+2008_001074
+2008_001075
+2008_001076
+2008_001077
+2008_001078
+2008_001080
+2008_001089
+2008_001090
+2008_001092
+2008_001098
+2008_001099
+2008_001111
+2008_001113
+2008_001114
+2008_001120
+2008_001121
+2008_001122
+2008_001135
+2008_001136
+2008_001139
+2008_001140
+2008_001142
+2008_001150
+2008_001154
+2008_001155
+2008_001160
+2008_001166
+2008_001167
+2008_001168
+2008_001170
+2008_001177
+2008_001183
+2008_001185
+2008_001192
+2008_001194
+2008_001199
+2008_001205
+2008_001210
+2008_001218
+2008_001220
+2008_001221
+2008_001225
+2008_001226
+2008_001227
+2008_001231
+2008_001236
+2008_001241
+2008_001248
+2008_001249
+2008_001255
+2008_001257
+2008_001260
+2008_001262
+2008_001264
+2008_001271
+2008_001275
+2008_001283
+2008_001284
+2008_001290
+2008_001296
+2008_001301
+2008_001304
+2008_001306
+2008_001308
+2008_001314
+2008_001318
+2008_001320
+2008_001322
+2008_001333
+2008_001334
+2008_001338
+2008_001340
+2008_001344
+2008_001349
+2008_001350
+2008_001353
+2008_001356
+2008_001366
+2008_001367
+2008_001369
+2008_001374
+2008_001376
+2008_001379
+2008_001380
+2008_001388
+2008_001391
+2008_001395
+2008_001401
+2008_001404
+2008_001406
+2008_001410
+2008_001415
+2008_001427
+2008_001428
+2008_001429
+2008_001430
+2008_001432
+2008_001433
+2008_001436
+2008_001437
+2008_001439
+2008_001445
+2008_001451
+2008_001456
+2008_001466
+2008_001468
+2008_001470
+2008_001475
+2008_001478
+2008_001481
+2008_001486
+2008_001491
+2008_001494
+2008_001503
+2008_001504
+2008_001513
+2008_001514
+2008_001516
+2008_001520
+2008_001522
+2008_001527
+2008_001531
+2008_001534
+2008_001536
+2008_001539
+2008_001540
+2008_001542
+2008_001543
+2008_001544
+2008_001546
+2008_001547
+2008_001549
+2008_001551
+2008_001553
+2008_001564
+2008_001574
+2008_001575
+2008_001580
+2008_001586
+2008_001589
+2008_001590
+2008_001593
+2008_001594
+2008_001596
+2008_001598
+2008_001602
+2008_001605
+2008_001607
+2008_001613
+2008_001619
+2008_001622
+2008_001624
+2008_001625
+2008_001629
+2008_001636
+2008_001638
+2008_001640
+2008_001648
+2008_001649
+2008_001655
+2008_001659
+2008_001660
+2008_001663
+2008_001667
+2008_001668
+2008_001669
+2008_001676
+2008_001680
+2008_001681
+2008_001682
+2008_001688
+2008_001697
+2008_001702
+2008_001712
+2008_001714
+2008_001715
+2008_001717
+2008_001722
+2008_001723
+2008_001727
+2008_001730
+2008_001731
+2008_001736
+2008_001742
+2008_001745
+2008_001750
+2008_001757
+2008_001763
+2008_001764
+2008_001765
+2008_001769
+2008_001772
+2008_001773
+2008_001774
+2008_001782
+2008_001784
+2008_001792
+2008_001799
+2008_001802
+2008_001805
+2008_001806
+2008_001808
+2008_001810
+2008_001812
+2008_001814
+2008_001815
+2008_001816
+2008_001820
+2008_001821
+2008_001823
+2008_001825
+2008_001830
+2008_001838
+2008_001841
+2008_001843
+2008_001850
+2008_001858
+2008_001862
+2008_001863
+2008_001867
+2008_001869
+2008_001871
+2008_001874
+2008_001885
+2008_001895
+2008_001899
+2008_001905
+2008_001907
+2008_001908
+2008_001909
+2008_001910
+2008_001914
+2008_001919
+2008_001920
+2008_001928
+2008_001930
+2008_001932
+2008_001934
+2008_001945
+2008_001946
+2008_001951
+2008_001958
+2008_001961
+2008_001965
+2008_001966
+2008_001969
+2008_001971
+2008_001978
+2008_001979
+2008_001985
+2008_001987
+2008_001989
+2008_001992
+2008_001998
+2008_002003
+2008_002004
+2008_002007
+2008_002011
+2008_002013
+2008_002017
+2008_002021
+2008_002026
+2008_002031
+2008_002035
+2008_002036
+2008_002037
+2008_002039
+2008_002042
+2008_002043
+2008_002045
+2008_002046
+2008_002047
+2008_002052
+2008_002069
+2008_002071
+2008_002082
+2008_002084
+2008_002086
+2008_002088
+2008_002092
+2008_002098
+2008_002099
+2008_002107
+2008_002113
+2008_002114
+2008_002115
+2008_002124
+2008_002132
+2008_002138
+2008_002140
+2008_002144
+2008_002146
+2008_002151
+2008_002152
+2008_002153
+2008_002155
+2008_002158
+2008_002167
+2008_002169
+2008_002172
+2008_002176
+2008_002179
+2008_002185
+2008_002191
+2008_002193
+2008_002194
+2008_002198
+2008_002199
+2008_002201
+2008_002205
+2008_002207
+2008_002209
+2008_002212
+2008_002222
+2008_002223
+2008_002231
+2008_002234
+2008_002239
+2008_002240
+2008_002241
+2008_002250
+2008_002259
+2008_002267
+2008_002269
+2008_002272
+2008_002273
+2008_002283
+2008_002292
+2008_002293
+2008_002298
+2008_002305
+2008_002312
+2008_002314
+2008_002321
+2008_002322
+2008_002324
+2008_002328
+2008_002330
+2008_002347
+2008_002349
+2008_002356
+2008_002358
+2008_002359
+2008_002366
+2008_002372
+2008_002374
+2008_002378
+2008_002379
+2008_002383
+2008_002384
+2008_002395
+2008_002401
+2008_002403
+2008_002404
+2008_002408
+2008_002410
+2008_002412
+2008_002414
+2008_002419
+2008_002424
+2008_002428
+2008_002429
+2008_002430
+2008_002436
+2008_002438
+2008_002439
+2008_002444
+2008_002445
+2008_002446
+2008_002451
+2008_002452
+2008_002454
+2008_002456
+2008_002464
+2008_002467
+2008_002470
+2008_002477
+2008_002481
+2008_002483
+2008_002485
+2008_002492
+2008_002494
+2008_002495
+2008_002499
+2008_002502
+2008_002504
+2008_002508
+2008_002509
+2008_002510
+2008_002512
+2008_002516
+2008_002521
+2008_002523
+2008_002526
+2008_002527
+2008_002536
+2008_002540
+2008_002542
+2008_002558
+2008_002564
+2008_002567
+2008_002576
+2008_002579
+2008_002588
+2008_002589
+2008_002590
+2008_002597
+2008_002598
+2008_002599
+2008_002603
+2008_002606
+2008_002610
+2008_002616
+2008_002623
+2008_002624
+2008_002631
+2008_002639
+2008_002640
+2008_002643
+2008_002645
+2008_002652
+2008_002653
+2008_002670
+2008_002673
+2008_002677
+2008_002678
+2008_002680
+2008_002681
+2008_002682
+2008_002684
+2008_002687
+2008_002696
+2008_002700
+2008_002701
+2008_002705
+2008_002709
+2008_002714
+2008_002715
+2008_002716
+2008_002720
+2008_002725
+2008_002732
+2008_002735
+2008_002738
+2008_002746
+2008_002751
+2008_002752
+2008_002753
+2008_002756
+2008_002766
+2008_002768
+2008_002773
+2008_002775
+2008_002778
+2008_002783
+2008_002789
+2008_002792
+2008_002795
+2008_002801
+2008_002806
+2008_002809
+2008_002811
+2008_002814
+2008_002817
+2008_002820
+2008_002826
+2008_002829
+2008_002830
+2008_002831
+2008_002835
+2008_002838
+2008_002843
+2008_002845
+2008_002847
+2008_002852
+2008_002859
+2008_002860
+2008_002864
+2008_002869
+2008_002870
+2008_002875
+2008_002876
+2008_002879
+2008_002882
+2008_002883
+2008_002897
+2008_002899
+2008_002900
+2008_002904
+2008_002906
+2008_002908
+2008_002909
+2008_002910
+2008_002920
+2008_002926
+2008_002929
+2008_002932
+2008_002936
+2008_002942
+2008_002946
+2008_002947
+2008_002956
+2008_002958
+2008_002965
+2008_002968
+2008_002971
+2008_002973
+2008_002992
+2008_002999
+2008_003001
+2008_003003
+2008_003005
+2008_003008
+2008_003020
+2008_003022
+2008_003026
+2008_003030
+2008_003034
+2008_003045
+2008_003051
+2008_003052
+2008_003053
+2008_003055
+2008_003056
+2008_003062
+2008_003067
+2008_003072
+2008_003073
+2008_003076
+2008_003082
+2008_003088
+2008_003089
+2008_003090
+2008_003095
+2008_003100
+2008_003104
+2008_003105
+2008_003106
+2008_003107
+2008_003108
+2008_003110
+2008_003132
+2008_003133
+2008_003135
+2008_003136
+2008_003141
+2008_003144
+2008_003152
+2008_003155
+2008_003161
+2008_003167
+2008_003170
+2008_003178
+2008_003181
+2008_003186
+2008_003187
+2008_003189
+2008_003193
+2008_003202
+2008_003205
+2008_003210
+2008_003211
+2008_003220
+2008_003222
+2008_003225
+2008_003228
+2008_003232
+2008_003238
+2008_003239
+2008_003245
+2008_003249
+2008_003255
+2008_003256
+2008_003263
+2008_003270
+2008_003271
+2008_003278
+2008_003280
+2008_003286
+2008_003289
+2008_003291
+2008_003295
+2008_003300
+2008_003305
+2008_003316
+2008_003320
+2008_003326
+2008_003330
+2008_003331
+2008_003333
+2008_003334
+2008_003336
+2008_003343
+2008_003344
+2008_003348
+2008_003350
+2008_003351
+2008_003359
+2008_003361
+2008_003369
+2008_003374
+2008_003379
+2008_003382
+2008_003384
+2008_003395
+2008_003402
+2008_003405
+2008_003407
+2008_003420
+2008_003423
+2008_003424
+2008_003432
+2008_003433
+2008_003439
+2008_003442
+2008_003443
+2008_003449
+2008_003451
+2008_003453
+2008_003461
+2008_003464
+2008_003466
+2008_003467
+2008_003472
+2008_003475
+2008_003476
+2008_003477
+2008_003479
+2008_003482
+2008_003483
+2008_003484
+2008_003492
+2008_003493
+2008_003499
+2008_003511
+2008_003514
+2008_003519
+2008_003524
+2008_003526
+2008_003531
+2008_003542
+2008_003545
+2008_003546
+2008_003547
+2008_003552
+2008_003557
+2008_003565
+2008_003572
+2008_003576
+2008_003577
+2008_003580
+2008_003591
+2008_003592
+2008_003593
+2008_003598
+2008_003604
+2008_003607
+2008_003609
+2008_003610
+2008_003613
+2008_003618
+2008_003619
+2008_003621
+2008_003624
+2008_003636
+2008_003638
+2008_003647
+2008_003650
+2008_003658
+2008_003662
+2008_003671
+2008_003672
+2008_003673
+2008_003676
+2008_003680
+2008_003681
+2008_003683
+2008_003684
+2008_003694
+2008_003704
+2008_003709
+2008_003713
+2008_003718
+2008_003720
+2008_003721
+2008_003722
+2008_003733
+2008_003737
+2008_003743
+2008_003744
+2008_003745
+2008_003749
+2008_003753
+2008_003754
+2008_003755
+2008_003756
+2008_003763
+2008_003766
+2008_003767
+2008_003768
+2008_003772
+2008_003775
+2008_003777
+2008_003780
+2008_003782
+2008_003789
+2008_003793
+2008_003794
+2008_003799
+2008_003800
+2008_003801
+2008_003805
+2008_003812
+2008_003813
+2008_003820
+2008_003821
+2008_003825
+2008_003826
+2008_003827
+2008_003829
+2008_003830
+2008_003835
+2008_003838
+2008_003840
+2008_003843
+2008_003844
+2008_003846
+2008_003856
+2008_003858
+2008_003860
+2008_003868
+2008_003873
+2008_003874
+2008_003876
+2008_003881
+2008_003884
+2008_003885
+2008_003886
+2008_003894
+2008_003904
+2008_003905
+2008_003915
+2008_003921
+2008_003924
+2008_003926
+2008_003929
+2008_003932
+2008_003933
+2008_003940
+2008_003941
+2008_003943
+2008_003944
+2008_003945
+2008_003948
+2008_003951
+2008_003958
+2008_003962
+2008_003965
+2008_003969
+2008_003971
+2008_003976
+2008_003983
+2008_003988
+2008_003989
+2008_003996
+2008_003997
+2008_004002
+2008_004003
+2008_004006
+2008_004007
+2008_004015
+2008_004016
+2008_004018
+2008_004020
+2008_004027
+2008_004030
+2008_004040
+2008_004045
+2008_004046
+2008_004048
+2008_004054
+2008_004056
+2008_004058
+2008_004064
+2008_004069
+2008_004071
+2008_004075
+2008_004076
+2008_004081
+2008_004088
+2008_004090
+2008_004093
+2008_004101
+2008_004103
+2008_004105
+2008_004119
+2008_004121
+2008_004123
+2008_004124
+2008_004125
+2008_004126
+2008_004127
+2008_004135
+2008_004137
+2008_004140
+2008_004142
+2008_004155
+2008_004166
+2008_004174
+2008_004175
+2008_004178
+2008_004182
+2008_004188
+2008_004189
+2008_004190
+2008_004198
+2008_004203
+2008_004205
+2008_004212
+2008_004213
+2008_004214
+2008_004216
+2008_004221
+2008_004230
+2008_004234
+2008_004242
+2008_004243
+2008_004245
+2008_004251
+2008_004257
+2008_004258
+2008_004263
+2008_004270
+2008_004271
+2008_004273
+2008_004278
+2008_004279
+2008_004289
+2008_004290
+2008_004292
+2008_004297
+2008_004306
+2008_004308
+2008_004312
+2008_004317
+2008_004318
+2008_004324
+2008_004326
+2008_004327
+2008_004330
+2008_004333
+2008_004339
+2008_004344
+2008_004345
+2008_004347
+2008_004348
+2008_004354
+2008_004357
+2008_004361
+2008_004363
+2008_004367
+2008_004371
+2008_004374
+2008_004384
+2008_004389
+2008_004391
+2008_004394
+2008_004396
+2008_004399
+2008_004402
+2008_004406
+2008_004408
+2008_004414
+2008_004417
+2008_004419
+2008_004422
+2008_004425
+2008_004426
+2008_004427
+2008_004431
+2008_004433
+2008_004438
+2008_004445
+2008_004453
+2008_004455
+2008_004459
+2008_004460
+2008_004464
+2008_004469
+2008_004470
+2008_004471
+2008_004476
+2008_004477
+2008_004478
+2008_004479
+2008_004482
+2008_004487
+2008_004497
+2008_004498
+2008_004501
+2008_004502
+2008_004504
+2008_004510
+2008_004520
+2008_004522
+2008_004525
+2008_004526
+2008_004528
+2008_004533
+2008_004534
+2008_004538
+2008_004540
+2008_004541
+2008_004546
+2008_004549
+2008_004550
+2008_004552
+2008_004553
+2008_004554
+2008_004564
+2008_004575
+2008_004579
+2008_004589
+2008_004592
+2008_004599
+2008_004605
+2008_004606
+2008_004610
+2008_004612
+2008_004613
+2008_004614
+2008_004615
+2008_004619
+2008_004621
+2008_004624
+2008_004630
+2008_004632
+2008_004633
+2008_004640
+2008_004646
+2008_004647
+2008_004653
+2008_004654
+2008_004656
+2008_004659
+2008_004662
+2008_004665
+2008_004670
+2008_004684
+2008_004687
+2008_004688
+2008_004689
+2008_004695
+2008_004696
+2008_004701
+2008_004702
+2008_004704
+2008_004705
+2008_004706
+2008_004711
+2008_004716
+2008_004718
+2008_004720
+2008_004722
+2008_004726
+2008_004729
+2008_004730
+2008_004736
+2008_004740
+2008_004742
+2008_004745
+2008_004754
+2008_004756
+2008_004758
+2008_004760
+2008_004766
+2008_004768
+2008_004771
+2008_004774
+2008_004778
+2008_004784
+2008_004794
+2008_004795
+2008_004797
+2008_004805
+2008_004812
+2008_004814
+2008_004819
+2008_004825
+2008_004832
+2008_004833
+2008_004837
+2008_004851
+2008_004852
+2008_004854
+2008_004862
+2008_004866
+2008_004873
+2008_004875
+2008_004881
+2008_004885
+2008_004887
+2008_004894
+2008_004896
+2008_004898
+2008_004900
+2008_004904
+2008_004907
+2008_004910
+2008_004921
+2008_004923
+2008_004926
+2008_004930
+2008_004933
+2008_004935
+2008_004940
+2008_004942
+2008_004948
+2008_004955
+2008_004967
+2008_004968
+2008_004974
+2008_004975
+2008_004979
+2008_004982
+2008_004984
+2008_004986
+2008_004995
+2008_005001
+2008_005003
+2008_005008
+2008_005010
+2008_005015
+2008_005023
+2008_005032
+2008_005035
+2008_005037
+2008_005043
+2008_005046
+2008_005049
+2008_005054
+2008_005057
+2008_005061
+2008_005063
+2008_005065
+2008_005068
+2008_005070
+2008_005072
+2008_005085
+2008_005089
+2008_005092
+2008_005096
+2008_005097
+2008_005098
+2008_005105
+2008_005107
+2008_005109
+2008_005110
+2008_005111
+2008_005114
+2008_005117
+2008_005123
+2008_005139
+2008_005140
+2008_005147
+2008_005151
+2008_005156
+2008_005160
+2008_005166
+2008_005167
+2008_005175
+2008_005181
+2008_005182
+2008_005183
+2008_005185
+2008_005190
+2008_005191
+2008_005194
+2008_005197
+2008_005204
+2008_005205
+2008_005208
+2008_005215
+2008_005217
+2008_005233
+2008_005235
+2008_005242
+2008_005243
+2008_005244
+2008_005245
+2008_005251
+2008_005252
+2008_005253
+2008_005254
+2008_005255
+2008_005257
+2008_005260
+2008_005261
+2008_005270
+2008_005272
+2008_005276
+2008_005277
+2008_005282
+2008_005288
+2008_005304
+2008_005309
+2008_005313
+2008_005316
+2008_005319
+2008_005323
+2008_005327
+2008_005335
+2008_005337
+2008_005338
+2008_005346
+2008_005347
+2008_005348
+2008_005356
+2008_005359
+2008_005360
+2008_005361
+2008_005369
+2008_005373
+2008_005374
+2008_005378
+2008_005379
+2008_005393
+2008_005398
+2008_005399
+2008_005404
+2008_005406
+2008_005417
+2008_005421
+2008_005422
+2008_005423
+2008_005427
+2008_005431
+2008_005436
+2008_005439
+2008_005444
+2008_005445
+2008_005446
+2008_005447
+2008_005455
+2008_005460
+2008_005467
+2008_005469
+2008_005472
+2008_005480
+2008_005485
+2008_005490
+2008_005498
+2008_005501
+2008_005504
+2008_005510
+2008_005511
+2008_005522
+2008_005525
+2008_005530
+2008_005534
+2008_005538
+2008_005544
+2008_005548
+2008_005550
+2008_005552
+2008_005553
+2008_005563
+2008_005564
+2008_005566
+2008_005573
+2008_005574
+2008_005582
+2008_005588
+2008_005599
+2008_005601
+2008_005608
+2008_005611
+2008_005612
+2008_005614
+2008_005627
+2008_005628
+2008_005631
+2008_005633
+2008_005635
+2008_005637
+2008_005638
+2008_005642
+2008_005643
+2008_005646
+2008_005649
+2008_005652
+2008_005657
+2008_005660
+2008_005663
+2008_005664
+2008_005676
+2008_005677
+2008_005680
+2008_005681
+2008_005685
+2008_005686
+2008_005687
+2008_005691
+2008_005695
+2008_005699
+2008_005701
+2008_005702
+2008_005703
+2008_005720
+2008_005721
+2008_005726
+2008_005727
+2008_005732
+2008_005734
+2008_005735
+2008_005738
+2008_005748
+2008_005750
+2008_005763
+2008_005764
+2008_005768
+2008_005774
+2008_005777
+2008_005779
+2008_005788
+2008_005790
+2008_005792
+2008_005796
+2008_005798
+2008_005801
+2008_005805
+2008_005808
+2008_005812
+2008_005816
+2008_005821
+2008_005825
+2008_005831
+2008_005838
+2008_005846
+2008_005847
+2008_005848
+2008_005857
+2008_005860
+2008_005863
+2008_005865
+2008_005869
+2008_005875
+2008_005877
+2008_005881
+2008_005883
+2008_005884
+2008_005889
+2008_005895
+2008_005898
+2008_005904
+2008_005907
+2008_005914
+2008_005915
+2008_005918
+2008_005924
+2008_005928
+2008_005933
+2008_005934
+2008_005936
+2008_005939
+2008_005943
+2008_005957
+2008_005962
+2008_005964
+2008_005970
+2008_005975
+2008_005977
+2008_005978
+2008_005980
+2008_005984
+2008_005987
+2008_005989
+2008_006002
+2008_006007
+2008_006008
+2008_006010
+2008_006017
+2008_006021
+2008_006024
+2008_006027
+2008_006028
+2008_006031
+2008_006034
+2008_006036
+2008_006037
+2008_006038
+2008_006041
+2008_006042
+2008_006045
+2008_006047
+2008_006050
+2008_006052
+2008_006055
+2008_006058
+2008_006059
+2008_006063
+2008_006068
+2008_006071
+2008_006072
+2008_006082
+2008_006087
+2008_006088
+2008_006094
+2008_006100
+2008_006104
+2008_006108
+2008_006109
+2008_006112
+2008_006113
+2008_006117
+2008_006120
+2008_006130
+2008_006143
+2008_006144
+2008_006147
+2008_006148
+2008_006151
+2008_006154
+2008_006159
+2008_006163
+2008_006166
+2008_006169
+2008_006175
+2008_006178
+2008_006179
+2008_006185
+2008_006190
+2008_006200
+2008_006203
+2008_006205
+2008_006207
+2008_006216
+2008_006218
+2008_006219
+2008_006222
+2008_006227
+2008_006229
+2008_006233
+2008_006239
+2008_006240
+2008_006254
+2008_006262
+2008_006267
+2008_006269
+2008_006275
+2008_006282
+2008_006285
+2008_006288
+2008_006290
+2008_006298
+2008_006303
+2008_006307
+2008_006310
+2008_006311
+2008_006316
+2008_006320
+2008_006325
+2008_006327
+2008_006330
+2008_006337
+2008_006341
+2008_006347
+2008_006356
+2008_006359
+2008_006362
+2008_006366
+2008_006368
+2008_006377
+2008_006382
+2008_006392
+2008_006394
+2008_006397
+2008_006403
+2008_006407
+2008_006408
+2008_006409
+2008_006416
+2008_006424
+2008_006425
+2008_006429
+2008_006438
+2008_006441
+2008_006449
+2008_006452
+2008_006458
+2008_006463
+2008_006467
+2008_006477
+2008_006480
+2008_006487
+2008_006489
+2008_006502
+2008_006503
+2008_006517
+2008_006519
+2008_006520
+2008_006523
+2008_006524
+2008_006526
+2008_006528
+2008_006530
+2008_006534
+2008_006547
+2008_006548
+2008_006553
+2008_006554
+2008_006568
+2008_006576
+2008_006587
+2008_006588
+2008_006591
+2008_006600
+2008_006604
+2008_006605
+2008_006609
+2008_006611
+2008_006614
+2008_006616
+2008_006617
+2008_006621
+2008_006631
+2008_006635
+2008_006642
+2008_006646
+2008_006656
+2008_006660
+2008_006662
+2008_006665
+2008_006671
+2008_006684
+2008_006686
+2008_006690
+2008_006694
+2008_006696
+2008_006701
+2008_006703
+2008_006708
+2008_006710
+2008_006714
+2008_006716
+2008_006722
+2008_006731
+2008_006732
+2008_006743
+2008_006746
+2008_006747
+2008_006752
+2008_006758
+2008_006765
+2008_006773
+2008_006774
+2008_006776
+2008_006777
+2008_006779
+2008_006781
+2008_006784
+2008_006792
+2008_006793
+2008_006796
+2008_006797
+2008_006798
+2008_006800
+2008_006811
+2008_006813
+2008_006815
+2008_006816
+2008_006817
+2008_006824
+2008_006825
+2008_006828
+2008_006831
+2008_006833
+2008_006835
+2008_006837
+2008_006839
+2008_006841
+2008_006844
+2008_006849
+2008_006855
+2008_006863
+2008_006870
+2008_006874
+2008_006880
+2008_006885
+2008_006887
+2008_006890
+2008_006892
+2008_006896
+2008_006900
+2008_006904
+2008_006907
+2008_006912
+2008_006924
+2008_006925
+2008_006939
+2008_006941
+2008_006944
+2008_006948
+2008_006949
+2008_006951
+2008_006952
+2008_006956
+2008_006959
+2008_006967
+2008_006968
+2008_006979
+2008_006980
+2008_006981
+2008_006986
+2008_006987
+2008_006989
+2008_006991
+2008_006997
+2008_006998
+2008_006999
+2008_007006
+2008_007010
+2008_007019
+2008_007021
+2008_007025
+2008_007031
+2008_007032
+2008_007034
+2008_007042
+2008_007048
+2008_007050
+2008_007056
+2008_007057
+2008_007059
+2008_007064
+2008_007067
+2008_007070
+2008_007084
+2008_007086
+2008_007091
+2008_007096
+2008_007103
+2008_007105
+2008_007108
+2008_007112
+2008_007114
+2008_007119
+2008_007120
+2008_007123
+2008_007130
+2008_007133
+2008_007134
+2008_007143
+2008_007163
+2008_007164
+2008_007166
+2008_007167
+2008_007171
+2008_007176
+2008_007181
+2008_007182
+2008_007184
+2008_007187
+2008_007188
+2008_007189
+2008_007190
+2008_007194
+2008_007195
+2008_007196
+2008_007207
+2008_007214
+2008_007216
+2008_007219
+2008_007222
+2008_007225
+2008_007227
+2008_007229
+2008_007231
+2008_007241
+2008_007247
+2008_007250
+2008_007256
+2008_007264
+2008_007266
+2008_007269
+2008_007273
+2008_007277
+2008_007279
+2008_007282
+2008_007285
+2008_007287
+2008_007293
+2008_007295
+2008_007305
+2008_007311
+2008_007314
+2008_007317
+2008_007319
+2008_007323
+2008_007324
+2008_007327
+2008_007332
+2008_007334
+2008_007336
+2008_007339
+2008_007344
+2008_007348
+2008_007350
+2008_007352
+2008_007358
+2008_007374
+2008_007378
+2008_007384
+2008_007389
+2008_007390
+2008_007392
+2008_007393
+2008_007398
+2008_007402
+2008_007403
+2008_007404
+2008_007409
+2008_007415
+2008_007417
+2008_007430
+2008_007431
+2008_007434
+2008_007435
+2008_007441
+2008_007446
+2008_007452
+2008_007455
+2008_007458
+2008_007459
+2008_007461
+2008_007466
+2008_007476
+2008_007478
+2008_007480
+2008_007488
+2008_007494
+2008_007497
+2008_007498
+2008_007501
+2008_007507
+2008_007513
+2008_007514
+2008_007521
+2008_007525
+2008_007527
+2008_007529
+2008_007531
+2008_007534
+2008_007536
+2008_007538
+2008_007548
+2008_007558
+2008_007561
+2008_007567
+2008_007574
+2008_007579
+2008_007583
+2008_007585
+2008_007586
+2008_007587
+2008_007591
+2008_007594
+2008_007595
+2008_007596
+2008_007599
+2008_007610
+2008_007612
+2008_007617
+2008_007618
+2008_007623
+2008_007632
+2008_007635
+2008_007643
+2008_007649
+2008_007656
+2008_007661
+2008_007662
+2008_007668
+2008_007669
+2008_007673
+2008_007676
+2008_007677
+2008_007685
+2008_007688
+2008_007690
+2008_007693
+2008_007694
+2008_007702
+2008_007704
+2008_007706
+2008_007714
+2008_007716
+2008_007719
+2008_007729
+2008_007733
+2008_007735
+2008_007736
+2008_007737
+2008_007738
+2008_007739
+2008_007741
+2008_007745
+2008_007749
+2008_007757
+2008_007760
+2008_007764
+2008_007766
+2008_007768
+2008_007787
+2008_007791
+2008_007793
+2008_007794
+2008_007797
+2008_007798
+2008_007804
+2008_007806
+2008_007811
+2008_007814
+2008_007816
+2008_007819
+2008_007823
+2008_007827
+2008_007828
+2008_007831
+2008_007836
+2008_007839
+2008_007841
+2008_007850
+2008_007853
+2008_007854
+2008_007855
+2008_007871
+2008_007872
+2008_007875
+2008_007884
+2008_007887
+2008_007888
+2008_007890
+2008_007893
+2008_007902
+2008_007914
+2008_007915
+2008_007917
+2008_007922
+2008_007923
+2008_007931
+2008_007932
+2008_007933
+2008_007935
+2008_007936
+2008_007942
+2008_007945
+2008_007948
+2008_007949
+2008_007954
+2008_007955
+2008_007964
+2008_007966
+2008_007970
+2008_007981
+2008_007986
+2008_007989
+2008_007993
+2008_007994
+2008_008001
+2008_008011
+2008_008022
+2008_008024
+2008_008025
+2008_008029
+2008_008040
+2008_008044
+2008_008050
+2008_008052
+2008_008053
+2008_008055
+2008_008057
+2008_008066
+2008_008069
+2008_008075
+2008_008084
+2008_008086
+2008_008091
+2008_008093
+2008_008096
+2008_008103
+2008_008105
+2008_008109
+2008_008113
+2008_008115
+2008_008120
+2008_008123
+2008_008127
+2008_008130
+2008_008131
+2008_008134
+2008_008141
+2008_008145
+2008_008146
+2008_008155
+2008_008175
+2008_008177
+2008_008179
+2008_008185
+2008_008190
+2008_008191
+2008_008192
+2008_008203
+2008_008208
+2008_008210
+2008_008217
+2008_008221
+2008_008224
+2008_008231
+2008_008232
+2008_008233
+2008_008234
+2008_008235
+2008_008241
+2008_008246
+2008_008252
+2008_008254
+2008_008257
+2008_008268
+2008_008269
+2008_008271
+2008_008272
+2008_008278
+2008_008279
+2008_008284
+2008_008292
+2008_008296
+2008_008297
+2008_008301
+2008_008302
+2008_008307
+2008_008310
+2008_008313
+2008_008314
+2008_008318
+2008_008320
+2008_008322
+2008_008331
+2008_008335
+2008_008336
+2008_008337
+2008_008341
+2008_008346
+2008_008354
+2008_008357
+2008_008359
+2008_008362
+2008_008365
+2008_008373
+2008_008376
+2008_008377
+2008_008379
+2008_008380
+2008_008387
+2008_008388
+2008_008392
+2008_008393
+2008_008395
+2008_008406
+2008_008421
+2008_008424
+2008_008429
+2008_008433
+2008_008434
+2008_008435
+2008_008437
+2008_008439
+2008_008443
+2008_008444
+2008_008446
+2008_008450
+2008_008453
+2008_008461
+2008_008466
+2008_008467
+2008_008469
+2008_008470
+2008_008474
+2008_008488
+2008_008500
+2008_008501
+2008_008506
+2008_008512
+2008_008519
+2008_008524
+2008_008527
+2008_008531
+2008_008536
+2008_008537
+2008_008538
+2008_008552
+2008_008554
+2008_008564
+2008_008570
+2008_008574
+2008_008585
+2008_008588
+2008_008595
+2008_008598
+2008_008606
+2008_008611
+2008_008613
+2008_008615
+2008_008617
+2008_008619
+2008_008621
+2008_008622
+2008_008627
+2008_008628
+2008_008629
+2008_008632
+2008_008636
+2008_008652
+2008_008658
+2008_008659
+2008_008662
+2008_008675
+2008_008676
+2008_008679
+2008_008682
+2008_008683
+2008_008684
+2008_008690
+2008_008695
+2008_008700
+2008_008705
+2008_008708
+2008_008711
+2008_008713
+2008_008718
+2008_008724
+2008_008726
+2008_008732
+2008_008739
+2008_008746
+2008_008751
+2008_008753
+2008_008755
+2008_008758
+2008_008765
+2008_008767
+2008_008772
+2009_000001
+2009_000002
+2009_000009
+2009_000011
+2009_000012
+2009_000013
+2009_000017
+2009_000022
+2009_000026
+2009_000032
+2009_000035
+2009_000037
+2009_000039
+2009_000041
+2009_000045
+2009_000051
+2009_000055
+2009_000060
+2009_000063
+2009_000066
+2009_000067
+2009_000068
+2009_000072
+2009_000074
+2009_000078
+2009_000080
+2009_000084
+2009_000087
+2009_000089
+2009_000090
+2009_000093
+2009_000096
+2009_000097
+2009_000102
+2009_000121
+2009_000124
+2009_000136
+2009_000142
+2009_000146
+2009_000149
+2009_000156
+2009_000157
+2009_000158
+2009_000165
+2009_000169
+2009_000171
+2009_000181
+2009_000182
+2009_000183
+2009_000184
+2009_000189
+2009_000192
+2009_000198
+2009_000199
+2009_000201
+2009_000205
+2009_000206
+2009_000212
+2009_000214
+2009_000216
+2009_000219
+2009_000225
+2009_000242
+2009_000244
+2009_000247
+2009_000249
+2009_000254
+2009_000257
+2009_000260
+2009_000276
+2009_000282
+2009_000283
+2009_000284
+2009_000286
+2009_000288
+2009_000291
+2009_000293
+2009_000297
+2009_000298
+2009_000300
+2009_000304
+2009_000305
+2009_000308
+2009_000309
+2009_000312
+2009_000316
+2009_000318
+2009_000321
+2009_000328
+2009_000330
+2009_000335
+2009_000337
+2009_000342
+2009_000351
+2009_000354
+2009_000356
+2009_000366
+2009_000370
+2009_000378
+2009_000387
+2009_000389
+2009_000391
+2009_000397
+2009_000398
+2009_000399
+2009_000402
+2009_000410
+2009_000411
+2009_000412
+2009_000414
+2009_000417
+2009_000418
+2009_000421
+2009_000422
+2009_000426
+2009_000430
+2009_000435
+2009_000440
+2009_000446
+2009_000453
+2009_000455
+2009_000456
+2009_000457
+2009_000461
+2009_000466
+2009_000469
+2009_000472
+2009_000483
+2009_000487
+2009_000488
+2009_000496
+2009_000499
+2009_000501
+2009_000511
+2009_000512
+2009_000513
+2009_000516
+2009_000519
+2009_000523
+2009_000526
+2009_000536
+2009_000542
+2009_000545
+2009_000549
+2009_000550
+2009_000552
+2009_000558
+2009_000559
+2009_000563
+2009_000566
+2009_000568
+2009_000573
+2009_000574
+2009_000590
+2009_000593
+2009_000597
+2009_000606
+2009_000608
+2009_000611
+2009_000614
+2009_000615
+2009_000619
+2009_000624
+2009_000625
+2009_000628
+2009_000631
+2009_000634
+2009_000637
+2009_000641
+2009_000647
+2009_000648
+2009_000658
+2009_000661
+2009_000664
+2009_000670
+2009_000674
+2009_000675
+2009_000676
+2009_000677
+2009_000681
+2009_000683
+2009_000689
+2009_000691
+2009_000702
+2009_000704
+2009_000705
+2009_000712
+2009_000716
+2009_000719
+2009_000723
+2009_000724
+2009_000725
+2009_000726
+2009_000727
+2009_000730
+2009_000731
+2009_000732
+2009_000734
+2009_000741
+2009_000742
+2009_000756
+2009_000758
+2009_000760
+2009_000762
+2009_000763
+2009_000771
+2009_000778
+2009_000779
+2009_000782
+2009_000783
+2009_000791
+2009_000811
+2009_000812
+2009_000817
+2009_000820
+2009_000821
+2009_000823
+2009_000824
+2009_000825
+2009_000828
+2009_000829
+2009_000830
+2009_000837
+2009_000839
+2009_000840
+2009_000843
+2009_000845
+2009_000846
+2009_000851
+2009_000852
+2009_000856
+2009_000858
+2009_000862
+2009_000865
+2009_000871
+2009_000879
+2009_000886
+2009_000890
+2009_000892
+2009_000896
+2009_000897
+2009_000898
+2009_000901
+2009_000904
+2009_000909
+2009_000919
+2009_000923
+2009_000924
+2009_000925
+2009_000928
+2009_000931
+2009_000934
+2009_000935
+2009_000939
+2009_000948
+2009_000954
+2009_000955
+2009_000958
+2009_000960
+2009_000964
+2009_000966
+2009_000971
+2009_000985
+2009_000989
+2009_000991
+2009_000992
+2009_000995
+2009_000998
+2009_001000
+2009_001006
+2009_001007
+2009_001008
+2009_001011
+2009_001016
+2009_001021
+2009_001024
+2009_001026
+2009_001028
+2009_001030
+2009_001038
+2009_001044
+2009_001054
+2009_001055
+2009_001057
+2009_001061
+2009_001066
+2009_001069
+2009_001075
+2009_001082
+2009_001083
+2009_001084
+2009_001090
+2009_001094
+2009_001097
+2009_001106
+2009_001108
+2009_001113
+2009_001118
+2009_001120
+2009_001121
+2009_001126
+2009_001128
+2009_001134
+2009_001139
+2009_001148
+2009_001155
+2009_001160
+2009_001164
+2009_001166
+2009_001181
+2009_001184
+2009_001194
+2009_001195
+2009_001196
+2009_001198
+2009_001207
+2009_001212
+2009_001215
+2009_001225
+2009_001227
+2009_001240
+2009_001242
+2009_001243
+2009_001245
+2009_001249
+2009_001252
+2009_001255
+2009_001257
+2009_001259
+2009_001266
+2009_001278
+2009_001279
+2009_001286
+2009_001288
+2009_001289
+2009_001299
+2009_001300
+2009_001309
+2009_001313
+2009_001314
+2009_001316
+2009_001320
+2009_001321
+2009_001322
+2009_001326
+2009_001332
+2009_001333
+2009_001343
+2009_001345
+2009_001348
+2009_001349
+2009_001350
+2009_001355
+2009_001361
+2009_001363
+2009_001366
+2009_001367
+2009_001370
+2009_001371
+2009_001384
+2009_001387
+2009_001391
+2009_001393
+2009_001397
+2009_001398
+2009_001406
+2009_001407
+2009_001409
+2009_001411
+2009_001413
+2009_001414
+2009_001417
+2009_001419
+2009_001426
+2009_001427
+2009_001431
+2009_001433
+2009_001437
+2009_001440
+2009_001447
+2009_001449
+2009_001456
+2009_001468
+2009_001470
+2009_001479
+2009_001484
+2009_001490
+2009_001498
+2009_001501
+2009_001505
+2009_001509
+2009_001518
+2009_001519
+2009_001521
+2009_001522
+2009_001526
+2009_001534
+2009_001535
+2009_001536
+2009_001539
+2009_001549
+2009_001554
+2009_001562
+2009_001565
+2009_001568
+2009_001575
+2009_001577
+2009_001581
+2009_001587
+2009_001591
+2009_001593
+2009_001594
+2009_001606
+2009_001607
+2009_001617
+2009_001618
+2009_001621
+2009_001623
+2009_001627
+2009_001631
+2009_001633
+2009_001635
+2009_001643
+2009_001644
+2009_001645
+2009_001646
+2009_001648
+2009_001653
+2009_001663
+2009_001667
+2009_001673
+2009_001675
+2009_001682
+2009_001683
+2009_001684
+2009_001687
+2009_001696
+2009_001699
+2009_001707
+2009_001709
+2009_001713
+2009_001718
+2009_001723
+2009_001731
+2009_001733
+2009_001738
+2009_001741
+2009_001743
+2009_001752
+2009_001754
+2009_001758
+2009_001759
+2009_001764
+2009_001765
+2009_001767
+2009_001768
+2009_001774
+2009_001775
+2009_001778
+2009_001780
+2009_001784
+2009_001794
+2009_001799
+2009_001804
+2009_001805
+2009_001810
+2009_001811
+2009_001816
+2009_001818
+2009_001820
+2009_001822
+2009_001823
+2009_001830
+2009_001833
+2009_001835
+2009_001839
+2009_001848
+2009_001850
+2009_001851
+2009_001852
+2009_001853
+2009_001854
+2009_001858
+2009_001864
+2009_001881
+2009_001890
+2009_001905
+2009_001906
+2009_001907
+2009_001909
+2009_001911
+2009_001915
+2009_001916
+2009_001929
+2009_001931
+2009_001940
+2009_001941
+2009_001945
+2009_001949
+2009_001965
+2009_001967
+2009_001971
+2009_001973
+2009_001976
+2009_001977
+2009_001979
+2009_001980
+2009_001984
+2009_001988
+2009_001991
+2009_002001
+2009_002002
+2009_002003
+2009_002008
+2009_002009
+2009_002011
+2009_002012
+2009_002024
+2009_002031
+2009_002035
+2009_002039
+2009_002042
+2009_002046
+2009_002047
+2009_002053
+2009_002055
+2009_002056
+2009_002058
+2009_002061
+2009_002073
+2009_002078
+2009_002082
+2009_002087
+2009_002094
+2009_002097
+2009_002110
+2009_002111
+2009_002122
+2009_002127
+2009_002128
+2009_002131
+2009_002136
+2009_002137
+2009_002139
+2009_002141
+2009_002144
+2009_002150
+2009_002155
+2009_002164
+2009_002165
+2009_002169
+2009_002171
+2009_002175
+2009_002177
+2009_002185
+2009_002191
+2009_002194
+2009_002199
+2009_002202
+2009_002205
+2009_002208
+2009_002211
+2009_002212
+2009_002215
+2009_002219
+2009_002221
+2009_002222
+2009_002226
+2009_002228
+2009_002230
+2009_002231
+2009_002232
+2009_002238
+2009_002239
+2009_002242
+2009_002252
+2009_002257
+2009_002265
+2009_002267
+2009_002268
+2009_002272
+2009_002282
+2009_002286
+2009_002291
+2009_002295
+2009_002302
+2009_002305
+2009_002306
+2009_002308
+2009_002317
+2009_002319
+2009_002320
+2009_002325
+2009_002328
+2009_002333
+2009_002335
+2009_002346
+2009_002349
+2009_002350
+2009_002360
+2009_002363
+2009_002366
+2009_002370
+2009_002372
+2009_002373
+2009_002374
+2009_002380
+2009_002382
+2009_002390
+2009_002393
+2009_002398
+2009_002399
+2009_002400
+2009_002401
+2009_002407
+2009_002414
+2009_002415
+2009_002420
+2009_002432
+2009_002433
+2009_002436
+2009_002441
+2009_002444
+2009_002445
+2009_002449
+2009_002453
+2009_002457
+2009_002464
+2009_002465
+2009_002470
+2009_002471
+2009_002474
+2009_002475
+2009_002476
+2009_002477
+2009_002487
+2009_002488
+2009_002499
+2009_002500
+2009_002510
+2009_002512
+2009_002515
+2009_002517
+2009_002518
+2009_002521
+2009_002524
+2009_002525
+2009_002527
+2009_002531
+2009_002532
+2009_002535
+2009_002537
+2009_002539
+2009_002546
+2009_002549
+2009_002552
+2009_002562
+2009_002563
+2009_002566
+2009_002568
+2009_002569
+2009_002570
+2009_002571
+2009_002573
+2009_002580
+2009_002584
+2009_002591
+2009_002592
+2009_002594
+2009_002604
+2009_002607
+2009_002608
+2009_002609
+2009_002614
+2009_002618
+2009_002632
+2009_002634
+2009_002635
+2009_002638
+2009_002645
+2009_002649
+2009_002651
+2009_002663
+2009_002665
+2009_002667
+2009_002668
+2009_002669
+2009_002670
+2009_002673
+2009_002675
+2009_002680
+2009_002681
+2009_002683
+2009_002684
+2009_002687
+2009_002698
+2009_002708
+2009_002711
+2009_002712
+2009_002717
+2009_002727
+2009_002732
+2009_002733
+2009_002739
+2009_002741
+2009_002743
+2009_002744
+2009_002749
+2009_002752
+2009_002753
+2009_002754
+2009_002755
+2009_002762
+2009_002765
+2009_002771
+2009_002772
+2009_002774
+2009_002777
+2009_002778
+2009_002779
+2009_002785
+2009_002790
+2009_002800
+2009_002803
+2009_002806
+2009_002807
+2009_002808
+2009_002809
+2009_002816
+2009_002830
+2009_002833
+2009_002836
+2009_002837
+2009_002838
+2009_002841
+2009_002856
+2009_002865
+2009_002876
+2009_002877
+2009_002882
+2009_002883
+2009_002887
+2009_002888
+2009_002894
+2009_002898
+2009_002902
+2009_002910
+2009_002918
+2009_002920
+2009_002925
+2009_002928
+2009_002936
+2009_002938
+2009_002940
+2009_002941
+2009_002960
+2009_002962
+2009_002967
+2009_002975
+2009_002977
+2009_002978
+2009_002982
+2009_002985
+2009_002986
+2009_002990
+2009_002995
+2009_002998
+2009_003003
+2009_003005
+2009_003010
+2009_003013
+2009_003018
+2009_003020
+2009_003022
+2009_003023
+2009_003031
+2009_003033
+2009_003043
+2009_003044
+2009_003052
+2009_003058
+2009_003059
+2009_003063
+2009_003065
+2009_003070
+2009_003071
+2009_003074
+2009_003076
+2009_003080
+2009_003083
+2009_003084
+2009_003089
+2009_003097
+2009_003098
+2009_003105
+2009_003110
+2009_003114
+2009_003122
+2009_003123
+2009_003125
+2009_003126
+2009_003128
+2009_003129
+2009_003130
+2009_003132
+2009_003136
+2009_003140
+2009_003143
+2009_003144
+2009_003150
+2009_003151
+2009_003153
+2009_003154
+2009_003181
+2009_003183
+2009_003185
+2009_003189
+2009_003191
+2009_003193
+2009_003194
+2009_003196
+2009_003198
+2009_003199
+2009_003201
+2009_003204
+2009_003212
+2009_003214
+2009_003217
+2009_003224
+2009_003230
+2009_003238
+2009_003241
+2009_003247
+2009_003251
+2009_003254
+2009_003255
+2009_003259
+2009_003262
+2009_003266
+2009_003269
+2009_003271
+2009_003273
+2009_003276
+2009_003278
+2009_003282
+2009_003284
+2009_003288
+2009_003294
+2009_003297
+2009_003299
+2009_003300
+2009_003301
+2009_003304
+2009_003305
+2009_003311
+2009_003312
+2009_003320
+2009_003323
+2009_003343
+2009_003346
+2009_003347
+2009_003348
+2009_003351
+2009_003372
+2009_003373
+2009_003375
+2009_003376
+2009_003378
+2009_003379
+2009_003380
+2009_003387
+2009_003394
+2009_003399
+2009_003400
+2009_003406
+2009_003409
+2009_003411
+2009_003415
+2009_003417
+2009_003422
+2009_003431
+2009_003433
+2009_003440
+2009_003441
+2009_003445
+2009_003450
+2009_003453
+2009_003456
+2009_003457
+2009_003460
+2009_003462
+2009_003466
+2009_003467
+2009_003469
+2009_003476
+2009_003481
+2009_003487
+2009_003491
+2009_003494
+2009_003498
+2009_003499
+2009_003500
+2009_003504
+2009_003507
+2009_003508
+2009_003509
+2009_003517
+2009_003521
+2009_003523
+2009_003528
+2009_003530
+2009_003537
+2009_003538
+2009_003542
+2009_003543
+2009_003544
+2009_003549
+2009_003551
+2009_003554
+2009_003560
+2009_003564
+2009_003565
+2009_003566
+2009_003569
+2009_003571
+2009_003576
+2009_003581
+2009_003588
+2009_003589
+2009_003592
+2009_003598
+2009_003606
+2009_003607
+2009_003612
+2009_003618
+2009_003626
+2009_003627
+2009_003633
+2009_003635
+2009_003637
+2009_003638
+2009_003640
+2009_003642
+2009_003650
+2009_003655
+2009_003656
+2009_003664
+2009_003666
+2009_003669
+2009_003671
+2009_003679
+2009_003686
+2009_003689
+2009_003696
+2009_003698
+2009_003703
+2009_003704
+2009_003707
+2009_003710
+2009_003713
+2009_003714
+2009_003718
+2009_003725
+2009_003726
+2009_003738
+2009_003747
+2009_003751
+2009_003756
+2009_003758
+2009_003759
+2009_003771
+2009_003773
+2009_003776
+2009_003781
+2009_003785
+2009_003795
+2009_003800
+2009_003802
+2009_003804
+2009_003805
+2009_003806
+2009_003810
+2009_003813
+2009_003814
+2009_003821
+2009_003822
+2009_003832
+2009_003835
+2009_003836
+2009_003840
+2009_003847
+2009_003849
+2009_003855
+2009_003857
+2009_003858
+2009_003863
+2009_003870
+2009_003874
+2009_003879
+2009_003884
+2009_003892
+2009_003895
+2009_003899
+2009_003901
+2009_003902
+2009_003903
+2009_003904
+2009_003905
+2009_003908
+2009_003911
+2009_003914
+2009_003916
+2009_003928
+2009_003929
+2009_003936
+2009_003938
+2009_003944
+2009_003947
+2009_003950
+2009_003951
+2009_003955
+2009_003962
+2009_003965
+2009_003969
+2009_003971
+2009_003973
+2009_003977
+2009_003982
+2009_003986
+2009_003991
+2009_003992
+2009_003995
+2009_004001
+2009_004004
+2009_004016
+2009_004019
+2009_004021
+2009_004022
+2009_004031
+2009_004032
+2009_004033
+2009_004034
+2009_004038
+2009_004040
+2009_004043
+2009_004044
+2009_004050
+2009_004051
+2009_004052
+2009_004062
+2009_004070
+2009_004072
+2009_004075
+2009_004076
+2009_004084
+2009_004085
+2009_004092
+2009_004093
+2009_004099
+2009_004102
+2009_004108
+2009_004111
+2009_004113
+2009_004124
+2009_004125
+2009_004126
+2009_004128
+2009_004129
+2009_004131
+2009_004138
+2009_004140
+2009_004141
+2009_004142
+2009_004148
+2009_004150
+2009_004152
+2009_004157
+2009_004161
+2009_004163
+2009_004164
+2009_004166
+2009_004170
+2009_004175
+2009_004188
+2009_004193
+2009_004197
+2009_004203
+2009_004205
+2009_004207
+2009_004210
+2009_004217
+2009_004221
+2009_004224
+2009_004232
+2009_004233
+2009_004241
+2009_004242
+2009_004243
+2009_004247
+2009_004248
+2009_004255
+2009_004258
+2009_004262
+2009_004263
+2009_004272
+2009_004273
+2009_004274
+2009_004277
+2009_004284
+2009_004291
+2009_004298
+2009_004300
+2009_004303
+2009_004307
+2009_004309
+2009_004324
+2009_004329
+2009_004332
+2009_004346
+2009_004350
+2009_004359
+2009_004361
+2009_004364
+2009_004366
+2009_004377
+2009_004382
+2009_004390
+2009_004402
+2009_004403
+2009_004404
+2009_004410
+2009_004411
+2009_004414
+2009_004419
+2009_004435
+2009_004436
+2009_004440
+2009_004448
+2009_004453
+2009_004455
+2009_004456
+2009_004457
+2009_004468
+2009_004471
+2009_004478
+2009_004483
+2009_004494
+2009_004496
+2009_004497
+2009_004499
+2009_004502
+2009_004504
+2009_004507
+2009_004509
+2009_004518
+2009_004524
+2009_004525
+2009_004529
+2009_004530
+2009_004532
+2009_004535
+2009_004536
+2009_004537
+2009_004540
+2009_004542
+2009_004543
+2009_004548
+2009_004551
+2009_004552
+2009_004556
+2009_004559
+2009_004567
+2009_004568
+2009_004579
+2009_004580
+2009_004581
+2009_004587
+2009_004588
+2009_004590
+2009_004592
+2009_004594
+2009_004601
+2009_004607
+2009_004614
+2009_004623
+2009_004624
+2009_004625
+2009_004629
+2009_004630
+2009_004634
+2009_004635
+2009_004645
+2009_004648
+2009_004653
+2009_004655
+2009_004664
+2009_004669
+2009_004670
+2009_004677
+2009_004679
+2009_004686
+2009_004687
+2009_004697
+2009_004706
+2009_004713
+2009_004716
+2009_004718
+2009_004720
+2009_004721
+2009_004730
+2009_004732
+2009_004738
+2009_004744
+2009_004746
+2009_004748
+2009_004749
+2009_004754
+2009_004758
+2009_004763
+2009_004765
+2009_004768
+2009_004769
+2009_004780
+2009_004781
+2009_004782
+2009_004784
+2009_004789
+2009_004796
+2009_004799
+2009_004801
+2009_004812
+2009_004820
+2009_004822
+2009_004823
+2009_004828
+2009_004834
+2009_004841
+2009_004845
+2009_004848
+2009_004849
+2009_004856
+2009_004857
+2009_004858
+2009_004859
+2009_004865
+2009_004867
+2009_004868
+2009_004869
+2009_004872
+2009_004876
+2009_004882
+2009_004885
+2009_004886
+2009_004889
+2009_004895
+2009_004897
+2009_004899
+2009_004902
+2009_004913
+2009_004917
+2009_004922
+2009_004929
+2009_004930
+2009_004933
+2009_004934
+2009_004940
+2009_004942
+2009_004946
+2009_004947
+2009_004956
+2009_004961
+2009_004969
+2009_004971
+2009_004974
+2009_004982
+2009_004987
+2009_004988
+2009_004993
+2009_004994
+2009_004996
+2009_005001
+2009_005005
+2009_005008
+2009_005019
+2009_005025
+2009_005033
+2009_005036
+2009_005038
+2009_005060
+2009_005061
+2009_005062
+2009_005064
+2009_005068
+2009_005073
+2009_005078
+2009_005079
+2009_005080
+2009_005082
+2009_005083
+2009_005086
+2009_005087
+2009_005089
+2009_005098
+2009_005103
+2009_005104
+2009_005111
+2009_005114
+2009_005119
+2009_005137
+2009_005140
+2009_005148
+2009_005149
+2009_005150
+2009_005152
+2009_005153
+2009_005156
+2009_005158
+2009_005161
+2009_005165
+2009_005171
+2009_005172
+2009_005178
+2009_005185
+2009_005189
+2009_005190
+2009_005193
+2009_005202
+2009_005203
+2009_005204
+2009_005205
+2009_005210
+2009_005211
+2009_005215
+2009_005216
+2009_005217
+2009_005219
+2009_005220
+2009_005221
+2009_005222
+2009_005225
+2009_005229
+2009_005231
+2009_005232
+2009_005239
+2009_005242
+2009_005257
+2009_005260
+2009_005262
+2009_005267
+2009_005268
+2009_005279
+2009_005286
+2009_005288
+2009_005292
+2009_005294
+2009_005299
+2009_005300
+2009_005302
+2009_005309
+2009_005310
+2010_000001
+2010_000003
+2010_000015
+2010_000020
+2010_000024
+2010_000027
+2010_000033
+2010_000035
+2010_000036
+2010_000038
+2010_000050
+2010_000053
+2010_000054
+2010_000065
+2010_000069
+2010_000072
+2010_000074
+2010_000083
+2010_000084
+2010_000085
+2010_000087
+2010_000088
+2010_000090
+2010_000095
+2010_000097
+2010_000098
+2010_000099
+2010_000110
+2010_000113
+2010_000118
+2010_000127
+2010_000139
+2010_000140
+2010_000145
+2010_000151
+2010_000159
+2010_000160
+2010_000162
+2010_000163
+2010_000170
+2010_000172
+2010_000174
+2010_000175
+2010_000178
+2010_000184
+2010_000193
+2010_000194
+2010_000196
+2010_000197
+2010_000199
+2010_000202
+2010_000211
+2010_000213
+2010_000216
+2010_000218
+2010_000238
+2010_000241
+2010_000246
+2010_000247
+2010_000254
+2010_000256
+2010_000260
+2010_000261
+2010_000262
+2010_000266
+2010_000272
+2010_000273
+2010_000279
+2010_000283
+2010_000284
+2010_000286
+2010_000291
+2010_000295
+2010_000308
+2010_000309
+2010_000312
+2010_000313
+2010_000317
+2010_000318
+2010_000321
+2010_000324
+2010_000325
+2010_000327
+2010_000330
+2010_000335
+2010_000336
+2010_000342
+2010_000344
+2010_000349
+2010_000351
+2010_000352
+2010_000358
+2010_000370
+2010_000372
+2010_000374
+2010_000375
+2010_000376
+2010_000379
+2010_000381
+2010_000382
+2010_000384
+2010_000390
+2010_000399
+2010_000401
+2010_000406
+2010_000409
+2010_000418
+2010_000422
+2010_000426
+2010_000427
+2010_000431
+2010_000433
+2010_000435
+2010_000442
+2010_000444
+2010_000446
+2010_000449
+2010_000456
+2010_000461
+2010_000462
+2010_000468
+2010_000470
+2010_000474
+2010_000475
+2010_000483
+2010_000485
+2010_000493
+2010_000497
+2010_000502
+2010_000506
+2010_000510
+2010_000515
+2010_000524
+2010_000526
+2010_000530
+2010_000536
+2010_000537
+2010_000541
+2010_000547
+2010_000548
+2010_000552
+2010_000553
+2010_000559
+2010_000562
+2010_000572
+2010_000573
+2010_000574
+2010_000582
+2010_000583
+2010_000586
+2010_000590
+2010_000602
+2010_000603
+2010_000604
+2010_000608
+2010_000617
+2010_000621
+2010_000622
+2010_000624
+2010_000628
+2010_000633
+2010_000635
+2010_000639
+2010_000646
+2010_000647
+2010_000655
+2010_000665
+2010_000666
+2010_000669
+2010_000679
+2010_000682
+2010_000683
+2010_000689
+2010_000690
+2010_000692
+2010_000695
+2010_000697
+2010_000705
+2010_000711
+2010_000712
+2010_000722
+2010_000724
+2010_000726
+2010_000727
+2010_000729
+2010_000731
+2010_000735
+2010_000737
+2010_000738
+2010_000743
+2010_000744
+2010_000749
+2010_000754
+2010_000759
+2010_000761
+2010_000764
+2010_000771
+2010_000778
+2010_000786
+2010_000788
+2010_000791
+2010_000792
+2010_000797
+2010_000802
+2010_000805
+2010_000811
+2010_000814
+2010_000821
+2010_000822
+2010_000828
+2010_000829
+2010_000830
+2010_000831
+2010_000836
+2010_000846
+2010_000865
+2010_000866
+2010_000870
+2010_000874
+2010_000875
+2010_000876
+2010_000883
+2010_000889
+2010_000893
+2010_000897
+2010_000898
+2010_000904
+2010_000906
+2010_000907
+2010_000915
+2010_000918
+2010_000923
+2010_000927
+2010_000928
+2010_000929
+2010_000931
+2010_000941
+2010_000944
+2010_000945
+2010_000947
+2010_000948
+2010_000952
+2010_000955
+2010_000956
+2010_000959
+2010_000961
+2010_000968
+2010_000973
+2010_000975
+2010_000981
+2010_000989
+2010_000993
+2010_000996
+2010_001000
+2010_001006
+2010_001008
+2010_001009
+2010_001010
+2010_001011
+2010_001016
+2010_001017
+2010_001021
+2010_001024
+2010_001030
+2010_001032
+2010_001036
+2010_001042
+2010_001049
+2010_001051
+2010_001052
+2010_001057
+2010_001061
+2010_001069
+2010_001070
+2010_001077
+2010_001079
+2010_001080
+2010_001082
+2010_001085
+2010_001089
+2010_001099
+2010_001104
+2010_001107
+2010_001109
+2010_001117
+2010_001119
+2010_001124
+2010_001125
+2010_001127
+2010_001130
+2010_001147
+2010_001149
+2010_001151
+2010_001158
+2010_001163
+2010_001164
+2010_001172
+2010_001174
+2010_001181
+2010_001188
+2010_001189
+2010_001192
+2010_001201
+2010_001204
+2010_001206
+2010_001214
+2010_001215
+2010_001216
+2010_001218
+2010_001219
+2010_001220
+2010_001229
+2010_001234
+2010_001241
+2010_001242
+2010_001246
+2010_001251
+2010_001256
+2010_001257
+2010_001263
+2010_001264
+2010_001270
+2010_001272
+2010_001286
+2010_001287
+2010_001291
+2010_001292
+2010_001293
+2010_001294
+2010_001301
+2010_001305
+2010_001313
+2010_001315
+2010_001321
+2010_001325
+2010_001326
+2010_001327
+2010_001331
+2010_001333
+2010_001343
+2010_001351
+2010_001355
+2010_001357
+2010_001364
+2010_001367
+2010_001376
+2010_001382
+2010_001394
+2010_001403
+2010_001405
+2010_001407
+2010_001411
+2010_001412
+2010_001417
+2010_001421
+2010_001426
+2010_001432
+2010_001439
+2010_001441
+2010_001448
+2010_001449
+2010_001451
+2010_001452
+2010_001453
+2010_001455
+2010_001461
+2010_001463
+2010_001468
+2010_001473
+2010_001479
+2010_001486
+2010_001497
+2010_001501
+2010_001502
+2010_001505
+2010_001516
+2010_001518
+2010_001520
+2010_001522
+2010_001525
+2010_001528
+2010_001534
+2010_001535
+2010_001536
+2010_001539
+2010_001540
+2010_001543
+2010_001544
+2010_001548
+2010_001553
+2010_001557
+2010_001563
+2010_001571
+2010_001574
+2010_001577
+2010_001579
+2010_001584
+2010_001586
+2010_001587
+2010_001601
+2010_001606
+2010_001614
+2010_001625
+2010_001633
+2010_001635
+2010_001636
+2010_001637
+2010_001640
+2010_001645
+2010_001646
+2010_001652
+2010_001656
+2010_001659
+2010_001668
+2010_001669
+2010_001671
+2010_001675
+2010_001679
+2010_001680
+2010_001682
+2010_001685
+2010_001690
+2010_001691
+2010_001692
+2010_001697
+2010_001699
+2010_001705
+2010_001712
+2010_001717
+2010_001720
+2010_001731
+2010_001734
+2010_001737
+2010_001739
+2010_001749
+2010_001752
+2010_001754
+2010_001757
+2010_001760
+2010_001763
+2010_001767
+2010_001768
+2010_001771
+2010_001773
+2010_001777
+2010_001783
+2010_001787
+2010_001788
+2010_001796
+2010_001803
+2010_001814
+2010_001819
+2010_001820
+2010_001821
+2010_001823
+2010_001827
+2010_001828
+2010_001829
+2010_001830
+2010_001836
+2010_001837
+2010_001838
+2010_001843
+2010_001845
+2010_001851
+2010_001857
+2010_001862
+2010_001863
+2010_001868
+2010_001869
+2010_001877
+2010_001891
+2010_001892
+2010_001893
+2010_001904
+2010_001907
+2010_001908
+2010_001913
+2010_001916
+2010_001918
+2010_001921
+2010_001927
+2010_001929
+2010_001937
+2010_001938
+2010_001950
+2010_001951
+2010_001954
+2010_001956
+2010_001962
+2010_001966
+2010_001967
+2010_001968
+2010_001986
+2010_001987
+2010_001988
+2010_001992
+2010_001995
+2010_001998
+2010_002000
+2010_002002
+2010_002005
+2010_002006
+2010_002017
+2010_002019
+2010_002022
+2010_002025
+2010_002029
+2010_002030
+2010_002040
+2010_002041
+2010_002046
+2010_002048
+2010_002050
+2010_002058
+2010_002060
+2010_002067
+2010_002073
+2010_002085
+2010_002086
+2010_002089
+2010_002094
+2010_002096
+2010_002098
+2010_002100
+2010_002102
+2010_002105
+2010_002106
+2010_002113
+2010_002117
+2010_002124
+2010_002128
+2010_002133
+2010_002137
+2010_002138
+2010_002142
+2010_002146
+2010_002147
+2010_002150
+2010_002161
+2010_002167
+2010_002172
+2010_002175
+2010_002181
+2010_002182
+2010_002183
+2010_002187
+2010_002192
+2010_002194
+2010_002195
+2010_002199
+2010_002200
+2010_002211
+2010_002213
+2010_002219
+2010_002223
+2010_002224
+2010_002228
+2010_002229
+2010_002232
+2010_002244
+2010_002245
+2010_002247
+2010_002251
+2010_002255
+2010_002261
+2010_002269
+2010_002271
+2010_002276
+2010_002279
+2010_002283
+2010_002287
+2010_002289
+2010_002294
+2010_002303
+2010_002305
+2010_002307
+2010_002310
+2010_002313
+2010_002315
+2010_002316
+2010_002319
+2010_002321
+2010_002326
+2010_002332
+2010_002336
+2010_002337
+2010_002340
+2010_002348
+2010_002354
+2010_002357
+2010_002361
+2010_002365
+2010_002366
+2010_002370
+2010_002372
+2010_002373
+2010_002383
+2010_002388
+2010_002390
+2010_002396
+2010_002398
+2010_002402
+2010_002405
+2010_002406
+2010_002408
+2010_002409
+2010_002420
+2010_002422
+2010_002427
+2010_002436
+2010_002446
+2010_002448
+2010_002449
+2010_002450
+2010_002458
+2010_002460
+2010_002461
+2010_002468
+2010_002479
+2010_002480
+2010_002482
+2010_002484
+2010_002504
+2010_002510
+2010_002512
+2010_002516
+2010_002518
+2010_002526
+2010_002531
+2010_002533
+2010_002534
+2010_002536
+2010_002538
+2010_002539
+2010_002542
+2010_002543
+2010_002546
+2010_002547
+2010_002561
+2010_002565
+2010_002569
+2010_002578
+2010_002579
+2010_002580
+2010_002586
+2010_002587
+2010_002597
+2010_002598
+2010_002601
+2010_002602
+2010_002603
+2010_002605
+2010_002621
+2010_002623
+2010_002629
+2010_002631
+2010_002632
+2010_002638
+2010_002639
+2010_002645
+2010_002652
+2010_002654
+2010_002660
+2010_002661
+2010_002666
+2010_002667
+2010_002668
+2010_002676
+2010_002678
+2010_002679
+2010_002682
+2010_002691
+2010_002693
+2010_002695
+2010_002701
+2010_002704
+2010_002705
+2010_002710
+2010_002713
+2010_002714
+2010_002716
+2010_002721
+2010_002723
+2010_002725
+2010_002728
+2010_002736
+2010_002737
+2010_002740
+2010_002741
+2010_002754
+2010_002758
+2010_002763
+2010_002767
+2010_002770
+2010_002771
+2010_002774
+2010_002775
+2010_002780
+2010_002783
+2010_002789
+2010_002790
+2010_002791
+2010_002792
+2010_002793
+2010_002803
+2010_002807
+2010_002808
+2010_002814
+2010_002817
+2010_002822
+2010_002824
+2010_002827
+2010_002840
+2010_002845
+2010_002853
+2010_002854
+2010_002858
+2010_002860
+2010_002864
+2010_002868
+2010_002871
+2010_002873
+2010_002876
+2010_002877
+2010_002879
+2010_002881
+2010_002887
+2010_002900
+2010_002902
+2010_002905
+2010_002914
+2010_002921
+2010_002924
+2010_002927
+2010_002929
+2010_002930
+2010_002939
+2010_002940
+2010_002954
+2010_002956
+2010_002958
+2010_002960
+2010_002963
+2010_002965
+2010_002972
+2010_002980
+2010_002985
+2010_002988
+2010_002991
+2010_002993
+2010_002995
+2010_003014
+2010_003015
+2010_003016
+2010_003019
+2010_003024
+2010_003040
+2010_003043
+2010_003051
+2010_003054
+2010_003060
+2010_003067
+2010_003071
+2010_003072
+2010_003074
+2010_003081
+2010_003082
+2010_003091
+2010_003092
+2010_003098
+2010_003102
+2010_003103
+2010_003107
+2010_003112
+2010_003120
+2010_003122
+2010_003123
+2010_003127
+2010_003129
+2010_003132
+2010_003133
+2010_003135
+2010_003139
+2010_003146
+2010_003147
+2010_003154
+2010_003156
+2010_003160
+2010_003168
+2010_003176
+2010_003183
+2010_003187
+2010_003190
+2010_003199
+2010_003200
+2010_003201
+2010_003207
+2010_003212
+2010_003214
+2010_003219
+2010_003220
+2010_003223
+2010_003231
+2010_003232
+2010_003233
+2010_003236
+2010_003239
+2010_003240
+2010_003244
+2010_003248
+2010_003249
+2010_003251
+2010_003253
+2010_003257
+2010_003260
+2010_003270
+2010_003275
+2010_003276
+2010_003278
+2010_003279
+2010_003285
+2010_003287
+2010_003293
+2010_003299
+2010_003302
+2010_003303
+2010_003314
+2010_003316
+2010_003321
+2010_003325
+2010_003326
+2010_003331
+2010_003335
+2010_003341
+2010_003358
+2010_003361
+2010_003362
+2010_003365
+2010_003366
+2010_003368
+2010_003375
+2010_003376
+2010_003379
+2010_003381
+2010_003385
+2010_003390
+2010_003397
+2010_003398
+2010_003401
+2010_003402
+2010_003409
+2010_003411
+2010_003418
+2010_003419
+2010_003427
+2010_003429
+2010_003446
+2010_003450
+2010_003451
+2010_003453
+2010_003458
+2010_003461
+2010_003465
+2010_003467
+2010_003468
+2010_003470
+2010_003473
+2010_003479
+2010_003482
+2010_003488
+2010_003490
+2010_003493
+2010_003495
+2010_003496
+2010_003497
+2010_003503
+2010_003506
+2010_003508
+2010_003514
+2010_003520
+2010_003522
+2010_003527
+2010_003531
+2010_003532
+2010_003537
+2010_003540
+2010_003541
+2010_003547
+2010_003559
+2010_003561
+2010_003562
+2010_003563
+2010_003568
+2010_003569
+2010_003573
+2010_003579
+2010_003585
+2010_003588
+2010_003594
+2010_003597
+2010_003603
+2010_003605
+2010_003609
+2010_003610
+2010_003613
+2010_003628
+2010_003630
+2010_003632
+2010_003640
+2010_003641
+2010_003645
+2010_003653
+2010_003655
+2010_003659
+2010_003664
+2010_003667
+2010_003673
+2010_003675
+2010_003679
+2010_003687
+2010_003688
+2010_003695
+2010_003701
+2010_003708
+2010_003709
+2010_003716
+2010_003723
+2010_003724
+2010_003728
+2010_003729
+2010_003730
+2010_003731
+2010_003735
+2010_003742
+2010_003744
+2010_003745
+2010_003746
+2010_003755
+2010_003757
+2010_003758
+2010_003761
+2010_003762
+2010_003764
+2010_003768
+2010_003771
+2010_003772
+2010_003774
+2010_003779
+2010_003781
+2010_003792
+2010_003800
+2010_003801
+2010_003805
+2010_003806
+2010_003807
+2010_003811
+2010_003813
+2010_003820
+2010_003823
+2010_003826
+2010_003828
+2010_003847
+2010_003848
+2010_003852
+2010_003854
+2010_003855
+2010_003857
+2010_003859
+2010_003861
+2010_003863
+2010_003878
+2010_003879
+2010_003890
+2010_003898
+2010_003912
+2010_003915
+2010_003919
+2010_003920
+2010_003928
+2010_003933
+2010_003936
+2010_003939
+2010_003942
+2010_003943
+2010_003944
+2010_003947
+2010_003955
+2010_003956
+2010_003961
+2010_003966
+2010_003970
+2010_003971
+2010_003976
+2010_003980
+2010_003981
+2010_003983
+2010_003988
+2010_003999
+2010_004006
+2010_004010
+2010_004021
+2010_004023
+2010_004026
+2010_004027
+2010_004031
+2010_004036
+2010_004037
+2010_004041
+2010_004042
+2010_004050
+2010_004054
+2010_004056
+2010_004063
+2010_004064
+2010_004067
+2010_004073
+2010_004088
+2010_004094
+2010_004095
+2010_004096
+2010_004102
+2010_004104
+2010_004105
+2010_004107
+2010_004120
+2010_004124
+2010_004125
+2010_004129
+2010_004137
+2010_004139
+2010_004140
+2010_004141
+2010_004143
+2010_004145
+2010_004149
+2010_004157
+2010_004161
+2010_004165
+2010_004173
+2010_004178
+2010_004179
+2010_004182
+2010_004184
+2010_004187
+2010_004188
+2010_004193
+2010_004201
+2010_004207
+2010_004208
+2010_004209
+2010_004211
+2010_004219
+2010_004224
+2010_004225
+2010_004226
+2010_004227
+2010_004228
+2010_004229
+2010_004230
+2010_004238
+2010_004253
+2010_004254
+2010_004257
+2010_004263
+2010_004278
+2010_004279
+2010_004280
+2010_004286
+2010_004290
+2010_004291
+2010_004297
+2010_004304
+2010_004312
+2010_004313
+2010_004314
+2010_004318
+2010_004320
+2010_004322
+2010_004335
+2010_004337
+2010_004339
+2010_004341
+2010_004345
+2010_004348
+2010_004350
+2010_004351
+2010_004352
+2010_004355
+2010_004362
+2010_004369
+2010_004374
+2010_004380
+2010_004382
+2010_004387
+2010_004390
+2010_004391
+2010_004400
+2010_004404
+2010_004409
+2010_004415
+2010_004417
+2010_004419
+2010_004420
+2010_004422
+2010_004425
+2010_004428
+2010_004431
+2010_004432
+2010_004439
+2010_004447
+2010_004455
+2010_004456
+2010_004457
+2010_004460
+2010_004461
+2010_004469
+2010_004472
+2010_004475
+2010_004479
+2010_004483
+2010_004484
+2010_004486
+2010_004488
+2010_004503
+2010_004505
+2010_004506
+2010_004509
+2010_004515
+2010_004519
+2010_004520
+2010_004529
+2010_004533
+2010_004536
+2010_004537
+2010_004542
+2010_004543
+2010_004545
+2010_004550
+2010_004551
+2010_004553
+2010_004554
+2010_004556
+2010_004557
+2010_004559
+2010_004567
+2010_004570
+2010_004584
+2010_004585
+2010_004586
+2010_004588
+2010_004596
+2010_004597
+2010_004608
+2010_004618
+2010_004624
+2010_004627
+2010_004628
+2010_004629
+2010_004634
+2010_004635
+2010_004637
+2010_004642
+2010_004654
+2010_004659
+2010_004661
+2010_004662
+2010_004667
+2010_004670
+2010_004672
+2010_004677
+2010_004679
+2010_004681
+2010_004686
+2010_004691
+2010_004692
+2010_004697
+2010_004714
+2010_004722
+2010_004733
+2010_004735
+2010_004743
+2010_004747
+2010_004748
+2010_004750
+2010_004753
+2010_004756
+2010_004757
+2010_004763
+2010_004768
+2010_004772
+2010_004775
+2010_004778
+2010_004779
+2010_004783
+2010_004785
+2010_004786
+2010_004789
+2010_004792
+2010_004795
+2010_004804
+2010_004809
+2010_004813
+2010_004815
+2010_004817
+2010_004821
+2010_004825
+2010_004828
+2010_004829
+2010_004830
+2010_004836
+2010_004849
+2010_004854
+2010_004856
+2010_004857
+2010_004861
+2010_004865
+2010_004866
+2010_004868
+2010_004877
+2010_004889
+2010_004891
+2010_004894
+2010_004901
+2010_004903
+2010_004906
+2010_004908
+2010_004909
+2010_004917
+2010_004919
+2010_004921
+2010_004930
+2010_004931
+2010_004941
+2010_004946
+2010_004951
+2010_004952
+2010_004954
+2010_004957
+2010_004967
+2010_004980
+2010_004982
+2010_004989
+2010_004992
+2010_004994
+2010_004998
+2010_005000
+2010_005005
+2010_005006
+2010_005008
+2010_005013
+2010_005021
+2010_005023
+2010_005026
+2010_005031
+2010_005035
+2010_005042
+2010_005044
+2010_005046
+2010_005048
+2010_005049
+2010_005052
+2010_005053
+2010_005059
+2010_005061
+2010_005063
+2010_005066
+2010_005075
+2010_005079
+2010_005082
+2010_005083
+2010_005087
+2010_005096
+2010_005107
+2010_005108
+2010_005109
+2010_005115
+2010_005116
+2010_005118
+2010_005120
+2010_005123
+2010_005130
+2010_005136
+2010_005138
+2010_005141
+2010_005143
+2010_005148
+2010_005152
+2010_005158
+2010_005159
+2010_005160
+2010_005164
+2010_005166
+2010_005167
+2010_005169
+2010_005174
+2010_005180
+2010_005184
+2010_005185
+2010_005187
+2010_005188
+2010_005192
+2010_005206
+2010_005208
+2010_005215
+2010_005222
+2010_005224
+2010_005226
+2010_005230
+2010_005239
+2010_005242
+2010_005243
+2010_005245
+2010_005246
+2010_005250
+2010_005252
+2010_005264
+2010_005268
+2010_005272
+2010_005284
+2010_005285
+2010_005287
+2010_005292
+2010_005293
+2010_005305
+2010_005314
+2010_005323
+2010_005327
+2010_005330
+2010_005331
+2010_005332
+2010_005338
+2010_005340
+2010_005344
+2010_005345
+2010_005346
+2010_005353
+2010_005366
+2010_005369
+2010_005372
+2010_005374
+2010_005375
+2010_005379
+2010_005382
+2010_005394
+2010_005398
+2010_005401
+2010_005405
+2010_005406
+2010_005410
+2010_005414
+2010_005416
+2010_005421
+2010_005424
+2010_005425
+2010_005428
+2010_005432
+2010_005433
+2010_005441
+2010_005448
+2010_005452
+2010_005455
+2010_005456
+2010_005463
+2010_005467
+2010_005472
+2010_005474
+2010_005480
+2010_005482
+2010_005483
+2010_005484
+2010_005491
+2010_005493
+2010_005496
+2010_005501
+2010_005502
+2010_005508
+2010_005514
+2010_005515
+2010_005516
+2010_005527
+2010_005531
+2010_005532
+2010_005534
+2010_005538
+2010_005542
+2010_005543
+2010_005548
+2010_005551
+2010_005556
+2010_005562
+2010_005566
+2010_005567
+2010_005572
+2010_005575
+2010_005576
+2010_005582
+2010_005586
+2010_005587
+2010_005592
+2010_005594
+2010_005606
+2010_005610
+2010_005612
+2010_005620
+2010_005625
+2010_005626
+2010_005632
+2010_005635
+2010_005636
+2010_005637
+2010_005644
+2010_005647
+2010_005651
+2010_005654
+2010_005657
+2010_005658
+2010_005664
+2010_005666
+2010_005671
+2010_005676
+2010_005681
+2010_005688
+2010_005692
+2010_005697
+2010_005705
+2010_005706
+2010_005709
+2010_005712
+2010_005718
+2010_005719
+2010_005727
+2010_005731
+2010_005733
+2010_005738
+2010_005747
+2010_005752
+2010_005754
+2010_005756
+2010_005761
+2010_005762
+2010_005763
+2010_005764
+2010_005767
+2010_005768
+2010_005777
+2010_005780
+2010_005784
+2010_005788
+2010_005804
+2010_005806
+2010_005815
+2010_005817
+2010_005824
+2010_005827
+2010_005833
+2010_005837
+2010_005838
+2010_005843
+2010_005848
+2010_005849
+2010_005853
+2010_005860
+2010_005868
+2010_005870
+2010_005871
+2010_005877
+2010_005882
+2010_005883
+2010_005884
+2010_005885
+2010_005886
+2010_005888
+2010_005894
+2010_005896
+2010_005897
+2010_005899
+2010_005901
+2010_005903
+2010_005907
+2010_005914
+2010_005922
+2010_005934
+2010_005936
+2010_005937
+2010_005938
+2010_005943
+2010_005953
+2010_005973
+2010_005976
+2010_005980
+2010_005981
+2010_005991
+2010_005992
+2010_005993
+2010_005997
+2010_005998
+2010_006000
+2010_006003
+2010_006004
+2010_006010
+2010_006011
+2010_006021
+2010_006025
+2010_006026
+2010_006031
+2010_006032
+2010_006033
+2010_006034
+2010_006035
+2010_006037
+2010_006041
+2010_006051
+2010_006054
+2010_006056
+2010_006057
+2010_006058
+2010_006061
+2010_006062
+2010_006070
+2010_006076
+2010_006082
+2010_006084
+2010_006086
+2011_000002
+2011_000007
+2011_000009
+2011_000010
+2011_000016
+2011_000034
+2011_000036
+2011_000037
+2011_000038
+2011_000043
+2011_000045
+2011_000051
+2011_000054
+2011_000057
+2011_000060
+2011_000061
+2011_000065
+2011_000066
+2011_000070
+2011_000071
+2011_000076
+2011_000077
+2011_000082
+2011_000083
+2011_000084
+2011_000086
+2011_000087
+2011_000090
+2011_000094
+2011_000096
+2011_000098
+2011_000102
+2011_000103
+2011_000109
+2011_000112
+2011_000114
+2011_000124
+2011_000128
+2011_000129
+2011_000130
+2011_000142
+2011_000146
+2011_000147
+2011_000161
+2011_000162
+2011_000163
+2011_000165
+2011_000166
+2011_000173
+2011_000178
+2011_000180
+2011_000185
+2011_000194
+2011_000195
+2011_000202
+2011_000206
+2011_000210
+2011_000213
+2011_000214
+2011_000226
+2011_000229
+2011_000232
+2011_000234
+2011_000238
+2011_000239
+2011_000246
+2011_000248
+2011_000253
+2011_000257
+2011_000273
+2011_000276
+2011_000283
+2011_000288
+2011_000291
+2011_000299
+2011_000304
+2011_000307
+2011_000309
+2011_000310
+2011_000312
+2011_000314
+2011_000315
+2011_000319
+2011_000320
+2011_000321
+2011_000322
+2011_000332
+2011_000338
+2011_000344
+2011_000346
+2011_000364
+2011_000369
+2011_000374
+2011_000386
+2011_000391
+2011_000396
+2011_000404
+2011_000408
+2011_000412
+2011_000418
+2011_000419
+2011_000426
+2011_000427
+2011_000432
+2011_000435
+2011_000436
+2011_000438
+2011_000445
+2011_000455
+2011_000456
+2011_000471
+2011_000474
+2011_000477
+2011_000479
+2011_000481
+2011_000482
+2011_000487
+2011_000498
+2011_000503
+2011_000511
+2011_000512
+2011_000514
+2011_000518
+2011_000519
+2011_000521
+2011_000526
+2011_000530
+2011_000532
+2011_000536
+2011_000541
+2011_000548
+2011_000554
+2011_000557
+2011_000559
+2011_000566
+2011_000569
+2011_000575
+2011_000585
+2011_000592
+2011_000598
+2011_000600
+2011_000607
+2011_000608
+2011_000609
+2011_000612
+2011_000618
+2011_000622
+2011_000627
+2011_000630
+2011_000634
+2011_000638
+2011_000656
+2011_000658
+2011_000661
+2011_000666
+2011_000669
+2011_000679
+2011_000683
+2011_000685
+2011_000688
+2011_000690
+2011_000709
+2011_000718
+2011_000724
+2011_000734
+2011_000743
+2011_000744
+2011_000745
+2011_000747
+2011_000749
+2011_000753
+2011_000765
+2011_000767
+2011_000770
+2011_000772
+2011_000774
+2011_000778
+2011_000780
+2011_000784
+2011_000785
+2011_000789
+2011_000807
+2011_000809
+2011_000813
+2011_000824
+2011_000830
+2011_000843
+2011_000850
+2011_000851
+2011_000853
+2011_000872
+2011_000874
+2011_000887
+2011_000888
+2011_000897
+2011_000900
+2011_000901
+2011_000908
+2011_000909
+2011_000912
+2011_000917
+2011_000919
+2011_000927
+2011_000930
+2011_000932
+2011_000933
+2011_000950
+2011_000951
+2011_000953
+2011_000957
+2011_000961
+2011_000965
+2011_000969
+2011_000977
+2011_000986
+2011_000990
+2011_001005
+2011_001008
+2011_001014
+2011_001019
+2011_001020
+2011_001025
+2011_001029
+2011_001032
+2011_001036
+2011_001040
+2011_001044
+2011_001047
+2011_001054
+2011_001056
+2011_001058
+2011_001060
+2011_001064
+2011_001069
+2011_001071
+2011_001081
+2011_001082
+2011_001084
+2011_001086
+2011_001100
+2011_001105
+2011_001106
+2011_001110
+2011_001111
+2011_001114
+2011_001116
+2011_001124
+2011_001126
+2011_001128
+2011_001137
+2011_001138
+2011_001146
+2011_001149
+2011_001152
+2011_001158
+2011_001159
+2011_001160
+2011_001161
+2011_001167
+2011_001190
+2011_001201
+2011_001203
+2011_001213
+2011_001217
+2011_001221
+2011_001223
+2011_001226
+2011_001229
+2011_001232
+2011_001245
+2011_001251
+2011_001252
+2011_001260
+2011_001261
+2011_001263
+2011_001264
+2011_001266
+2011_001271
+2011_001276
+2011_001281
+2011_001282
+2011_001283
+2011_001284
+2011_001287
+2011_001288
+2011_001290
+2011_001292
+2011_001295
+2011_001304
+2011_001305
+2011_001311
+2011_001313
+2011_001315
+2011_001319
+2011_001326
+2011_001327
+2011_001329
+2011_001330
+2011_001335
+2011_001337
+2011_001341
+2011_001346
+2011_001350
+2011_001355
+2011_001360
+2011_001366
+2011_001370
+2011_001375
+2011_001387
+2011_001388
+2011_001389
+2011_001390
+2011_001399
+2011_001404
+2011_001406
+2011_001407
+2011_001416
+2011_001421
+2011_001434
+2011_001440
+2011_001441
+2011_001447
+2011_001467
+2011_001471
+2011_001489
+2011_001501
+2011_001507
+2011_001508
+2011_001518
+2011_001521
+2011_001524
+2011_001525
+2011_001529
+2011_001530
+2011_001531
+2011_001534
+2011_001535
+2011_001541
+2011_001543
+2011_001544
+2011_001546
+2011_001558
+2011_001567
+2011_001568
+2011_001573
+2011_001589
+2011_001591
+2011_001592
+2011_001596
+2011_001597
+2011_001601
+2011_001607
+2011_001608
+2011_001612
+2011_001613
+2011_001614
+2011_001618
+2011_001619
+2011_001620
+2011_001624
+2011_001628
+2011_001641
+2011_001642
+2011_001655
+2011_001665
+2011_001669
+2011_001674
+2011_001678
+2011_001691
+2011_001693
+2011_001699
+2011_001705
+2011_001707
+2011_001708
+2011_001712
+2011_001713
+2011_001714
+2011_001719
+2011_001720
+2011_001722
+2011_001726
+2011_001741
+2011_001745
+2011_001747
+2011_001748
+2011_001751
+2011_001757
+2011_001770
+2011_001771
+2011_001775
+2011_001782
+2011_001785
+2011_001793
+2011_001794
+2011_001800
+2011_001801
+2011_001806
+2011_001812
+2011_001815
+2011_001819
+2011_001820
+2011_001822
+2011_001824
+2011_001825
+2011_001827
+2011_001834
+2011_001837
+2011_001841
+2011_001842
+2011_001845
+2011_001847
+2011_001854
+2011_001856
+2011_001858
+2011_001862
+2011_001863
+2011_001868
+2011_001870
+2011_001873
+2011_001876
+2011_001877
+2011_001880
+2011_001900
+2011_001910
+2011_001911
+2011_001914
+2011_001919
+2011_001927
+2011_001932
+2011_001941
+2011_001942
+2011_001945
+2011_001946
+2011_001951
+2011_001962
+2011_001966
+2011_001975
+2011_001980
+2011_001982
+2011_001984
+2011_001986
+2011_001988
+2011_001989
+2011_002002
+2011_002003
+2011_002004
+2011_002016
+2011_002018
+2011_002019
+2011_002021
+2011_002033
+2011_002036
+2011_002038
+2011_002040
+2011_002041
+2011_002042
+2011_002044
+2011_002045
+2011_002047
+2011_002064
+2011_002074
+2011_002075
+2011_002079
+2011_002088
+2011_002091
+2011_002093
+2011_002098
+2011_002100
+2011_002102
+2011_002105
+2011_002108
+2011_002109
+2011_002110
+2011_002116
+2011_002121
+2011_002124
+2011_002128
+2011_002132
+2011_002137
+2011_002150
+2011_002154
+2011_002156
+2011_002158
+2011_002159
+2011_002160
+2011_002163
+2011_002169
+2011_002173
+2011_002174
+2011_002178
+2011_002184
+2011_002185
+2011_002192
+2011_002193
+2011_002200
+2011_002215
+2011_002221
+2011_002223
+2011_002230
+2011_002234
+2011_002241
+2011_002244
+2011_002247
+2011_002248
+2011_002260
+2011_002269
+2011_002270
+2011_002272
+2011_002276
+2011_002279
+2011_002280
+2011_002292
+2011_002294
+2011_002295
+2011_002298
+2011_002301
+2011_002308
+2011_002312
+2011_002317
+2011_002322
+2011_002324
+2011_002325
+2011_002327
+2011_002330
+2011_002343
+2011_002357
+2011_002358
+2011_002362
+2011_002365
+2011_002366
+2011_002371
+2011_002379
+2011_002380
+2011_002384
+2011_002386
+2011_002391
+2011_002393
+2011_002395
+2011_002396
+2011_002406
+2011_002407
+2011_002409
+2011_002414
+2011_002429
+2011_002453
+2011_002459
+2011_002463
+2011_002479
+2011_002482
+2011_002490
+2011_002491
+2011_002494
+2011_002495
+2011_002498
+2011_002505
+2011_002507
+2011_002509
+2011_002515
+2011_002516
+2011_002519
+2011_002520
+2011_002531
+2011_002532
+2011_002535
+2011_002536
+2011_002542
+2011_002548
+2011_002556
+2011_002558
+2011_002566
+2011_002575
+2011_002578
+2011_002579
+2011_002582
+2011_002583
+2011_002588
+2011_002589
+2011_002592
+2011_002605
+2011_002610
+2011_002612
+2011_002617
+2011_002623
+2011_002624
+2011_002629
+2011_002631
+2011_002639
+2011_002640
+2011_002641
+2011_002644
+2011_002662
+2011_002674
+2011_002675
+2011_002678
+2011_002685
+2011_002687
+2011_002694
+2011_002699
+2011_002713
+2011_002714
+2011_002725
+2011_002730
+2011_002738
+2011_002740
+2011_002742
+2011_002750
+2011_002751
+2011_002754
+2011_002760
+2011_002765
+2011_002766
+2011_002772
+2011_002784
+2011_002786
+2011_002796
+2011_002802
+2011_002805
+2011_002810
+2011_002812
+2011_002817
+2011_002830
+2011_002831
+2011_002833
+2011_002838
+2011_002841
+2011_002854
+2011_002863
+2011_002864
+2011_002868
+2011_002870
+2011_002871
+2011_002879
+2011_002880
+2011_002883
+2011_002885
+2011_002887
+2011_002890
+2011_002897
+2011_002900
+2011_002916
+2011_002925
+2011_002929
+2011_002933
+2011_002943
+2011_002944
+2011_002951
+2011_002962
+2011_002967
+2011_002970
+2011_002971
+2011_002975
+2011_002978
+2011_002983
+2011_002985
+2011_002992
+2011_002993
+2011_002994
+2011_002997
+2011_002999
+2011_003003
+2011_003011
+2011_003012
+2011_003013
+2011_003019
+2011_003023
+2011_003027
+2011_003028
+2011_003029
+2011_003030
+2011_003039
+2011_003043
+2011_003050
+2011_003055
+2011_003059
+2011_003076
+2011_003079
+2011_003085
+2011_003086
+2011_003089
+2011_003097
+2011_003098
+2011_003103
+2011_003111
+2011_003114
+2011_003115
+2011_003145
+2011_003146
+2011_003149
+2011_003152
+2011_003163
+2011_003166
+2011_003167
+2011_003168
+2011_003169
+2011_003176
+2011_003182
+2011_003185
+2011_003197
+2011_003201
+2011_003205
+2011_003207
+2011_003211
+2011_003212
+2011_003213
+2011_003220
+2011_003228
+2011_003232
+2011_003240
+2011_003242
+2011_003244
+2011_003254
+2011_003256
+2011_003260
+2011_003261
+2011_003262
+2011_003269
+2011_003271
+2011_003275
diff --git a/ImageSets/README.md b/ImageSets/README.md
new file mode 100644
index 0000000..44b3a3a
--- /dev/null
+++ b/ImageSets/README.md
@@ -0,0 +1 @@
+Please copy this folder to `$RFCN_ROOT/data/VOCdevkit0712/VOC0712`
diff --git a/LICENSE b/LICENSE
new file mode 100644
index 0000000..317ecb6
--- /dev/null
+++ b/LICENSE
@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2016 Yuwen Xiong
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..f3601c0
--- /dev/null
+++ b/README.md
@@ -0,0 +1,150 @@
+# py-R-FCN
+R-FCN: Object Detection via Region-based Fully Convolutional Networks
+
+py-R-FCN now supports end-to-end training. 
+
+### Disclaimer
+
+The official R-FCN code (written in MATLAB) is available [here](https://github.com/daijifeng001/R-FCN).
+If your goal is to reproduce the results in the [NIPS 2016 paper](https://arxiv.org/abs/1605.06409), please use the [official code](https://github.com/daijifeng001/R-FCN).
+
+py-R-FCN is based on the [py-faster-rcnn code](https://github.com/rbgirshick/py-faster-rcnn ) and [the offcial R-FCN implementation](https://github.com/daijifeng001/R-FCN), and the usage is quite similar to [py-faster-rcnn](https://github.com/rbgirshick/py-faster-rcnn ), thanks for them.
+
+There are slight differences between the two implementations.
+In particular, this Python port
+ - is ~10% slower at test-time, because some operations execute on the CPU in Python layers (e.g., 90ms / image vs. 99ms / image for ResNet-50)
+ - gives similar, but not exactly the same, mAP as the MATLAB version
+
+The original py-faster-rcnn uses class-aware bounding box regression. However, R-FCN use agonistic bounding box regression to reduce model complexity. So I add a configuration AGONISTIC into fast_rcnn/config.py, and the default value is False. You should set it to True both on train and test phase if you want to use class agonistic training and test.
+
+### License
+
+R-FCN is released under the MIT License (refer to the LICENSE file for details).
+
+### Citing R-FCN
+
+If you find R-FCN useful in your research, please consider citing:
+
+    @article{dai16rfcn,
+        Author = {Jifeng Dai, Yi Li, Kaiming He, Jian Sun},
+        Title = {{R-FCN}: Object Detection via Region-based Fully Convolutional Networks},
+        Journal = {arXiv preprint arXiv:1605.06409},
+        Year = {2016}
+    }
+    
+### Main Results
+                   | training data       | test data             | mAP   | time/img (K40) | time/img (Titian X)
+-------------------|:-------------------:|:---------------------:|:-----:|:--------------:|:------------------:|
+R-FCN, ResNet-50  | VOC 07+12 trainval  | VOC 07 test           | 76.8% | N/A        | 0.099sec            |
+R-FCN, ResNet-101 | VOC 07+12 trainval  | VOC 07 test           | N/A | N/A        | N/A            |
+
+
+### Requirements: software
+
+0. **`Important`** Please use the [Microsoft-version Caffe(@commit 1a2be8e)](https://github.com/Microsoft/caffe/tree/1a2be8ecf9ba318d516d79187845e90ac6e73197), this Caffe supports R-FCN layer, and the prototxt in this repository follows the Microsoft-version Caffe's layer name. You need to put the Caffe root folder under py-R-FCN folder, just like what py-faster-rcnn does.
+
+1. Requirements for `Caffe` and `pycaffe` (see: [Caffe installation instructions](http://caffe.berkeleyvision.org/installation.html))
+
+  **Note:** Caffe *must* be built with support for Python layers!
+
+  ```make
+  # In your Makefile.config, make sure to have this line uncommented
+  WITH_PYTHON_LAYER := 1
+  # Unrelatedly, it's also recommended that you use CUDNN
+  USE_CUDNN := 1
+  ```
+2. Python packages you might not have: `cython`, `python-opencv`, `easydict`
+3. [Optional] MATLAB is required for **official** PASCAL VOC evaluation only. The code now includes unofficial Python evaluation code.
+
+### Requirements: hardware
+
+Any NVIDIA GPU with 6GB or larger memory is OK(4GB is enough for ResNet-50).
+
+### Demo
+1.  I do not provide demo currently, I'll add it soon.
+
+### Preparation for Training & Testing
+1. Download the training, validation, test data and VOCdevkit
+
+	```Shell
+	wget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtrainval_06-Nov-2007.tar
+	wget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtest_06-Nov-2007.tar
+	wget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCdevkit_08-Jun-2007.tar
+	wget http://host.robots.ox.ac.uk/pascal/VOC/voc2012/VOCtrainval_11-May-2012.tar
+	```
+
+2. Extract all of these tars into one directory named `VOCdevkit`
+
+	```Shell
+	tar xvf VOCtrainval_06-Nov-2007.tar
+	tar xvf VOCtest_06-Nov-2007.tar
+	tar xvf VOCdevkit_08-Jun-2007.tar
+	tar xvf VOCtrainval_11-May-2012.tar
+	```
+
+3. It should have this basic structure
+
+	```Shell
+  	$VOCdevkit/                           # development kit
+  	$VOCdevkit/VOCcode/                   # VOC utility code
+  	$VOCdevkit/VOC2007                    # image sets, annotations, etc.
+  	$VOCdevkit/VOC2012                    # image sets, annotations, etc.
+  	# ... and several other directories ...
+  	```
+
+4. Since py-faster-rcnn does not support muliple training dataset, we need to merge VOC 2007 data and VOC 2012 data maually. Just make a new directory named `VOC0712`, put all subfolder except `ImageSets` in `VOC2007` and `VOC2012` into `VOC0712`(you'll merge some folder). I provided merged-version ImageSet text files for you, please put them into VOCdevkit/VOC0712/ImageSets.
+
+5. Then the folder structure should look like this
+  ```Shell
+  	$VOCdevkit/                           # development kit
+  	$VOCdevkit/VOCcode/                   # VOC utility code
+  	$VOCdevkit/VOC2007                    # image sets, annotations, etc.
+  	$VOCdevkit/VOC2012                    # image sets, annotations, etc.
+  	$VOCdevkit/VOC0712                    # you just created this folder
+  	# ... and several other directories ...
+  ```
+
+4. Create symlinks for the PASCAL VOC dataset
+
+	```Shell
+    cd $RFCN_ROOT/data
+    ln -s $VOCdevkit VOCdevkit0712
+    ```
+
+5.  Please download ImageNet-pre-trained ResNet-50 and ResNet-100 model manually, and put them into `$RFCN_ROOT/data/imagenet_models`
+8.  Then everything is done, you could train your own model.
+
+### Usage
+
+To train and test a R-FCN detector using the **approximate joint training** method, use `experiments/scripts/rfcn_end2end.sh`.
+Output is written underneath `$RFCN_ROOT/output`.
+
+To train and test a R-FCN detector using the **approximate joint training** method **with OHEM**, use `experiments/scripts/rfcn_end2end_ohem.sh`.
+Output is written underneath `$RFCN_ROOT/output`.
+
+```Shell
+cd $RFCN_ROOT
+./experiments/scripts/rfcn_end2end[_ohem].sh [GPU_ID] [NET] [--set ...]
+# GPU_ID is the GPU you want to train on
+# NET in {ResNet-50, ResNet-101} is the network arch to use
+# --set ... allows you to specify fast_rcnn.config options, e.g.
+#   --set EXP_DIR seed_rng1701 RNG_SEED 1701
+```
+
+Trained R-FCN networks are saved under:
+
+```
+output/<experiment directory>/<dataset name>/
+```
+
+Test outputs are saved under:
+
+```
+output/<experiment directory>/<dataset name>/<network snapshot name>/
+```
+
+### Misc
+
+Tested on Ubuntu 14.04 with a Titan X GPU and Intel Xeon CPU E5-2620 v2 @ 2.10GHz 
+
+py-faster-rcnn code can also work properly, but I do not add any other feature(such as ResNet and OHEM).
diff --git a/data/.gitignore b/data/.gitignore
new file mode 100644
index 0000000..dd09a04
--- /dev/null
+++ b/data/.gitignore
@@ -0,0 +1,6 @@
+selective_search*
+imagenet_models*
+fast_rcnn_models*
+faster_rcnn_models*
+VOCdevkit*
+cache
diff --git a/data/README.md b/data/README.md
new file mode 100644
index 0000000..109c453
--- /dev/null
+++ b/data/README.md
@@ -0,0 +1,69 @@
+This directory holds (*after you download them*):
+- Caffe models pre-trained on ImageNet
+- Faster R-CNN models
+- Symlinks to datasets
+
+To download Caffe models (ZF, VGG16) pre-trained on ImageNet, run:
+
+```
+./data/scripts/fetch_imagenet_models.sh
+```
+
+This script will populate `data/imagenet_models`.
+
+To download Faster R-CNN models trained on VOC 2007, run:
+
+```
+./data/scripts/fetch_faster_rcnn_models.sh
+```
+
+This script will populate `data/faster_rcnn_models`.
+
+In order to train and test with PASCAL VOC, you will need to establish symlinks.
+From the `data` directory (`cd data`):
+
+```
+# For VOC 2007
+ln -s /your/path/to/VOC2007/VOCdevkit VOCdevkit2007
+
+# For VOC 2012
+ln -s /your/path/to/VOC2012/VOCdevkit VOCdevkit2012
+```
+
+Install the MS COCO dataset at /path/to/coco
+
+```
+ln -s /path/to/coco coco
+```
+
+For COCO with Fast R-CNN, place object proposals under `coco_proposals` (inside
+the `data` directory). You can obtain proposals on COCO from Jan Hosang at
+https://www.mpi-inf.mpg.de/departments/computer-vision-and-multimodal-computing/research/object-recognition-and-scene-understanding/how-good-are-detection-proposals-really/.
+For COCO, using MCG is recommended over selective search. MCG boxes can be downloaded
+from http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/mcg/.
+Use the tool `lib/datasets/tools/mcg_munge.py` to convert the downloaded MCG data
+into the same file layout as those from Jan Hosang.
+
+Since you'll likely be experimenting with multiple installs of Fast/er R-CNN in
+parallel, you'll probably want to keep all of this data in a shared place and
+use symlinks. On my system I create the following symlinks inside `data`:
+
+Annotations for the 5k image 'minival' subset of COCO val2014 that I like to use
+can be found at http://www.cs.berkeley.edu/~rbg/faster-rcnn-data/instances_minival2014.json.zip.
+Annotations for COCO val2014 (set) minus minival (~35k images) can be found at
+http://www.cs.berkeley.edu/~rbg/faster-rcnn-data/instances_valminusminival2014.json.zip.
+
+```
+# data/cache holds various outputs created by the datasets package
+ln -s /data/fast_rcnn_shared/cache
+
+# move the imagenet_models to shared location and symlink to them
+ln -s /data/fast_rcnn_shared/imagenet_models
+
+# move the selective search data to a shared location and symlink to them
+# (only applicable to Fast R-CNN training)
+ln -s /data/fast_rcnn_shared/selective_search_data
+
+ln -s /data/VOC2007/VOCdevkit VOCdevkit2007
+ln -s /data/VOC2012/VOCdevkit VOCdevkit2012
+```
diff --git a/data/demo/000456.jpg b/data/demo/000456.jpg
new file mode 100644
index 0000000..480afa5
Binary files /dev/null and b/data/demo/000456.jpg differ
diff --git a/data/demo/000542.jpg b/data/demo/000542.jpg
new file mode 100644
index 0000000..d87e919
Binary files /dev/null and b/data/demo/000542.jpg differ
diff --git a/data/demo/001150.jpg b/data/demo/001150.jpg
new file mode 100644
index 0000000..e8c86a4
Binary files /dev/null and b/data/demo/001150.jpg differ
diff --git a/data/demo/001763.jpg b/data/demo/001763.jpg
new file mode 100644
index 0000000..c80ebd9
Binary files /dev/null and b/data/demo/001763.jpg differ
diff --git a/data/demo/004545.jpg b/data/demo/004545.jpg
new file mode 100644
index 0000000..4e06c20
Binary files /dev/null and b/data/demo/004545.jpg differ
diff --git a/data/pylintrc b/data/pylintrc
new file mode 100644
index 0000000..a348aea
--- /dev/null
+++ b/data/pylintrc
@@ -0,0 +1,3 @@
+[TYPECHECK]
+
+ignored-modules = numpy, numpy.random, cv2
diff --git a/data/scripts/fetch_faster_rcnn_models.sh b/data/scripts/fetch_faster_rcnn_models.sh
new file mode 100755
index 0000000..019771d
--- /dev/null
+++ b/data/scripts/fetch_faster_rcnn_models.sh
@@ -0,0 +1,34 @@
+#!/bin/bash
+
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )/../" && pwd )"
+cd $DIR
+
+FILE=faster_rcnn_models.tgz
+URL=http://www.cs.berkeley.edu/~rbg/faster-rcnn-data/$FILE
+CHECKSUM=ac116844f66aefe29587214272054668
+
+if [ -f $FILE ]; then
+  echo "File already exists. Checking md5..."
+  os=`uname -s`
+  if [ "$os" = "Linux" ]; then
+    checksum=`md5sum $FILE | awk '{ print $1 }'`
+  elif [ "$os" = "Darwin" ]; then
+    checksum=`cat $FILE | md5`
+  fi
+  if [ "$checksum" = "$CHECKSUM" ]; then
+    echo "Checksum is correct. No need to download."
+    exit 0
+  else
+    echo "Checksum is incorrect. Need to download again."
+  fi
+fi
+
+echo "Downloading Faster R-CNN demo models (695M)..."
+
+wget $URL -O $FILE
+
+echo "Unzipping..."
+
+tar zxvf $FILE
+
+echo "Done. Please run this command again to verify that checksum = $CHECKSUM."
diff --git a/data/scripts/fetch_imagenet_models.sh b/data/scripts/fetch_imagenet_models.sh
new file mode 100755
index 0000000..de18050
--- /dev/null
+++ b/data/scripts/fetch_imagenet_models.sh
@@ -0,0 +1,34 @@
+#!/bin/bash
+
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )/../" && pwd )"
+cd $DIR
+
+FILE=imagenet_models.tgz
+URL=http://www.cs.berkeley.edu/~rbg/faster-rcnn-data/$FILE
+CHECKSUM=ed34ca912d6782edfb673a8c3a0bda6d
+
+if [ -f $FILE ]; then
+  echo "File already exists. Checking md5..."
+  os=`uname -s`
+  if [ "$os" = "Linux" ]; then
+    checksum=`md5sum $FILE | awk '{ print $1 }'`
+  elif [ "$os" = "Darwin" ]; then
+    checksum=`cat $FILE | md5`
+  fi
+  if [ "$checksum" = "$CHECKSUM" ]; then
+    echo "Checksum is correct. No need to download."
+    exit 0
+  else
+    echo "Checksum is incorrect. Need to download again."
+  fi
+fi
+
+echo "Downloading pretrained ImageNet models (1G)..."
+
+wget $URL -O $FILE
+
+echo "Unzipping..."
+
+tar zxvf $FILE
+
+echo "Done. Please run this command again to verify that checksum = $CHECKSUM."
diff --git a/data/scripts/fetch_selective_search_data.sh b/data/scripts/fetch_selective_search_data.sh
new file mode 100755
index 0000000..84c0aca
--- /dev/null
+++ b/data/scripts/fetch_selective_search_data.sh
@@ -0,0 +1,34 @@
+#!/bin/bash
+
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )/../" && pwd )"
+cd $DIR
+
+FILE=selective_search_data.tgz
+URL=http://www.cs.berkeley.edu/~rbg/fast-rcnn-data/$FILE
+CHECKSUM=7078c1db87a7851b31966b96774cd9b9
+
+if [ -f $FILE ]; then
+  echo "File already exists. Checking md5..."
+  os=`uname -s`
+  if [ "$os" = "Linux" ]; then
+    checksum=`md5sum $FILE | awk '{ print $1 }'`
+  elif [ "$os" = "Darwin" ]; then
+    checksum=`cat $FILE | md5`
+  fi
+  if [ "$checksum" = "$CHECKSUM" ]; then
+    echo "Checksum is correct. No need to download."
+    exit 0
+  else
+    echo "Checksum is incorrect. Need to download again."
+  fi
+fi
+
+echo "Downloading precomputed selective search boxes (0.5G)..."
+
+wget $URL -O $FILE
+
+echo "Unzipping..."
+
+tar zxvf $FILE
+
+echo "Done. Please run this command again to verify that checksum = $CHECKSUM."
diff --git a/experiments/README.md b/experiments/README.md
new file mode 100644
index 0000000..076257d
--- /dev/null
+++ b/experiments/README.md
@@ -0,0 +1,5 @@
+Scripts are under `experiments/scripts`.
+
+Each script saves a log file under `experiments/logs`.
+
+Configuration override files used in the experiments are stored in `experiments/cfgs`.
diff --git a/experiments/cfgs/faster_rcnn_alt_opt.yml b/experiments/cfgs/faster_rcnn_alt_opt.yml
new file mode 100644
index 0000000..9f11bd6
--- /dev/null
+++ b/experiments/cfgs/faster_rcnn_alt_opt.yml
@@ -0,0 +1,5 @@
+EXP_DIR: faster_rcnn_alt_opt
+TRAIN:
+  BG_THRESH_LO: 0.0
+TEST:
+  HAS_RPN: True
diff --git a/experiments/cfgs/faster_rcnn_end2end.yml b/experiments/cfgs/faster_rcnn_end2end.yml
new file mode 100644
index 0000000..4c6a054
--- /dev/null
+++ b/experiments/cfgs/faster_rcnn_end2end.yml
@@ -0,0 +1,11 @@
+EXP_DIR: faster_rcnn_end2end
+TRAIN:
+  HAS_RPN: True
+  IMS_PER_BATCH: 1
+  BBOX_NORMALIZE_TARGETS_PRECOMPUTED: True
+  RPN_POSITIVE_OVERLAP: 0.7
+  RPN_BATCHSIZE: 256
+  PROPOSAL_METHOD: gt
+  BG_THRESH_LO: 0.0
+TEST:
+  HAS_RPN: True
diff --git a/experiments/cfgs/rfcn_end2end.yml b/experiments/cfgs/rfcn_end2end.yml
new file mode 100644
index 0000000..ebeca75
--- /dev/null
+++ b/experiments/cfgs/rfcn_end2end.yml
@@ -0,0 +1,15 @@
+EXP_DIR: rfcn_end2end
+TRAIN:
+  HAS_RPN: True
+  IMS_PER_BATCH: 1
+  BBOX_NORMALIZE_TARGETS_PRECOMPUTED: True
+  RPN_POSITIVE_OVERLAP: 0.7
+  RPN_BATCHSIZE: 256
+  PROPOSAL_METHOD: gt
+  BG_THRESH_LO: 0.1
+  BATCH_SIZE: 128
+  AGONISTIC: True
+  SNAPSHOT_ITERS: 10000
+TEST:
+  HAS_RPN: True
+  AGONISTIC: True
diff --git a/experiments/cfgs/rfcn_end2end_ohem.yml b/experiments/cfgs/rfcn_end2end_ohem.yml
new file mode 100644
index 0000000..dc00697
--- /dev/null
+++ b/experiments/cfgs/rfcn_end2end_ohem.yml
@@ -0,0 +1,16 @@
+EXP_DIR: rfcn_end2end_ohem
+TRAIN:
+  HAS_RPN: True
+  IMS_PER_BATCH: 1
+  BBOX_NORMALIZE_TARGETS_PRECOMPUTED: True
+  RPN_POSITIVE_OVERLAP: 0.7
+  RPN_BATCHSIZE: 256
+  PROPOSAL_METHOD: gt
+  BG_THRESH_LO: 0.0
+  BATCH_SIZE: -1
+  AGONISTIC: True
+  SNAPSHOT_ITERS: 1000
+  RPN_MIN_SIZE: 16 
+TEST:
+  HAS_RPN: True
+  AGONISTIC: True
diff --git a/experiments/logs/.gitignore b/experiments/logs/.gitignore
new file mode 100644
index 0000000..355c023
--- /dev/null
+++ b/experiments/logs/.gitignore
@@ -0,0 +1 @@
+*.txt*
diff --git a/experiments/scripts/fast_rcnn.sh b/experiments/scripts/fast_rcnn.sh
new file mode 100755
index 0000000..9f4e72e
--- /dev/null
+++ b/experiments/scripts/fast_rcnn.sh
@@ -0,0 +1,63 @@
+#!/bin/bash
+# Usage:
+# ./experiments/scripts/fast_rcnn.sh GPU NET DATASET [options args to {train,test}_net.py]
+# DATASET is either pascal_voc or coco.
+#
+# Example:
+# ./experiments/scripts/fast_rcnn.sh 0 VGG_CNN_M_1024 pascal_voc \
+#   --set EXP_DIR foobar RNG_SEED 42 TRAIN.SCALES "[400, 500, 600, 700]"
+
+set -x
+set -e
+
+export PYTHONUNBUFFERED="True"
+
+GPU_ID=$1
+NET=$2
+NET_lc=${NET,,}
+DATASET=$3
+
+array=( $@ )
+len=${#array[@]}
+EXTRA_ARGS=${array[@]:3:$len}
+EXTRA_ARGS_SLUG=${EXTRA_ARGS// /_}
+
+case $DATASET in
+  pascal_voc)
+    TRAIN_IMDB="voc_2007_trainval"
+    TEST_IMDB="voc_2007_test"
+    PT_DIR="pascal_voc"
+    ITERS=40000
+    ;;
+  coco)
+    TRAIN_IMDB="coco_2014_train"
+    TEST_IMDB="coco_2014_minival"
+    PT_DIR="coco"
+    ITERS=280000
+    ;;
+  *)
+    echo "No dataset given"
+    exit
+    ;;
+esac
+
+LOG="experiments/logs/fast_rcnn_${NET}_${EXTRA_ARGS_SLUG}.txt.`date +'%Y-%m-%d_%H-%M-%S'`"
+exec &> >(tee -a "$LOG")
+echo Logging output to "$LOG"
+
+time ./tools/train_net.py --gpu ${GPU_ID} \
+  --solver models/${PT_DIR}/${NET}/fast_rcnn/solver.prototxt \
+  --weights data/imagenet_models/${NET}.v2.caffemodel \
+  --imdb ${TRAIN_IMDB} \
+  --iters ${ITERS} \
+  ${EXTRA_ARGS}
+
+set +x
+NET_FINAL=`grep -B 1 "done solving" ${LOG} | grep "Wrote snapshot" | awk '{print $4}'`
+set -x
+
+time ./tools/test_net.py --gpu ${GPU_ID} \
+  --def models/${PT_DIR}/${NET}/fast_rcnn/test.prototxt \
+  --net ${NET_FINAL} \
+  --imdb ${TEST_IMDB} \
+  ${EXTRA_ARGS}
diff --git a/experiments/scripts/faster_rcnn_alt_opt.sh b/experiments/scripts/faster_rcnn_alt_opt.sh
new file mode 100755
index 0000000..d2a1ebe
--- /dev/null
+++ b/experiments/scripts/faster_rcnn_alt_opt.sh
@@ -0,0 +1,62 @@
+#!/bin/bash
+# Usage:
+# ./experiments/scripts/faster_rcnn_alt_opt.sh GPU NET DATASET [options args to {train,test}_net.py]
+# DATASET is only pascal_voc for now
+#
+# Example:
+# ./experiments/scripts/faster_rcnn_alt_opt.sh 0 VGG_CNN_M_1024 pascal_voc \
+#   --set EXP_DIR foobar RNG_SEED 42 TRAIN.SCALES "[400, 500, 600, 700]"
+
+set -x
+set -e
+
+export PYTHONUNBUFFERED="True"
+
+GPU_ID=$1
+NET=$2
+NET_lc=${NET,,}
+DATASET=$3
+
+array=( $@ )
+len=${#array[@]}
+EXTRA_ARGS=${array[@]:3:$len}
+EXTRA_ARGS_SLUG=${EXTRA_ARGS// /_}
+
+case $DATASET in
+  pascal_voc)
+    TRAIN_IMDB="voc_2007_trainval"
+    TEST_IMDB="voc_2007_test"
+    PT_DIR="pascal_voc"
+    ITERS=40000
+    ;;
+  coco)
+    echo "Not implemented: use experiments/scripts/faster_rcnn_end2end.sh for coco"
+    exit
+    ;;
+  *)
+    echo "No dataset given"
+    exit
+    ;;
+esac
+
+LOG="experiments/logs/faster_rcnn_alt_opt_${NET}_${EXTRA_ARGS_SLUG}.txt.`date +'%Y-%m-%d_%H-%M-%S'`"
+exec &> >(tee -a "$LOG")
+echo Logging output to "$LOG"
+
+time ./tools/train_faster_rcnn_alt_opt.py --gpu ${GPU_ID} \
+  --net_name ${NET} \
+  --weights data/imagenet_models/${NET}.v2.caffemodel \
+  --imdb ${TRAIN_IMDB} \
+  --cfg experiments/cfgs/faster_rcnn_alt_opt.yml \
+  ${EXTRA_ARGS}
+
+set +x
+NET_FINAL=`grep "Final model:" ${LOG} | awk '{print $3}'`
+set -x
+
+time ./tools/test_net.py --gpu ${GPU_ID} \
+  --def models/${PT_DIR}/${NET}/faster_rcnn_alt_opt/faster_rcnn_test.pt \
+  --net ${NET_FINAL} \
+  --imdb ${TEST_IMDB} \
+  --cfg experiments/cfgs/faster_rcnn_alt_opt.yml \
+  ${EXTRA_ARGS}
diff --git a/experiments/scripts/faster_rcnn_end2end.sh b/experiments/scripts/faster_rcnn_end2end.sh
new file mode 100755
index 0000000..79770aa
--- /dev/null
+++ b/experiments/scripts/faster_rcnn_end2end.sh
@@ -0,0 +1,68 @@
+#!/bin/bash
+# Usage:
+# ./experiments/scripts/faster_rcnn_end2end.sh GPU NET DATASET [options args to {train,test}_net.py]
+# DATASET is either pascal_voc or coco.
+#
+# Example:
+# ./experiments/scripts/faster_rcnn_end2end.sh 0 VGG_CNN_M_1024 pascal_voc \
+#   --set EXP_DIR foobar RNG_SEED 42 TRAIN.SCALES "[400, 500, 600, 700]"
+
+set -x
+set -e
+
+export PYTHONUNBUFFERED="True"
+
+GPU_ID=$1
+NET=$2
+NET_lc=${NET,,}
+DATASET=$3
+
+array=( $@ )
+len=${#array[@]}
+EXTRA_ARGS=${array[@]:3:$len}
+EXTRA_ARGS_SLUG=${EXTRA_ARGS// /_}
+
+case $DATASET in
+  pascal_voc)
+    TRAIN_IMDB="voc_2007_trainval"
+    TEST_IMDB="voc_2007_test"
+    PT_DIR="pascal_voc"
+    ITERS=70000
+    ;;
+  coco)
+    # This is a very long and slow training schedule
+    # You can probably use fewer iterations and reduce the
+    # time to the LR drop (set in the solver to 350,000 iterations).
+    TRAIN_IMDB="coco_2014_train"
+    TEST_IMDB="coco_2014_minival"
+    PT_DIR="coco"
+    ITERS=490000
+    ;;
+  *)
+    echo "No dataset given"
+    exit
+    ;;
+esac
+
+LOG="experiments/logs/faster_rcnn_end2end_${NET}_${EXTRA_ARGS_SLUG}.txt.`date +'%Y-%m-%d_%H-%M-%S'`"
+exec &> >(tee -a "$LOG")
+echo Logging output to "$LOG"
+
+time ./tools/train_net.py --gpu ${GPU_ID} \
+  --solver models/${PT_DIR}/${NET}/faster_rcnn_end2end/solver.prototxt \
+  --weights data/imagenet_models/${NET}.v2.caffemodel \
+  --imdb ${TRAIN_IMDB} \
+  --iters ${ITERS} \
+  --cfg experiments/cfgs/faster_rcnn_end2end.yml \
+  ${EXTRA_ARGS}
+
+set +x
+NET_FINAL=`grep -B 1 "done solving" ${LOG} | grep "Wrote snapshot" | awk '{print $4}'`
+set -x
+
+time ./tools/test_net.py --gpu ${GPU_ID} \
+  --def models/${PT_DIR}/${NET}/faster_rcnn_end2end/test.prototxt \
+  --net ${NET_FINAL} \
+  --imdb ${TEST_IMDB} \
+  --cfg experiments/cfgs/faster_rcnn_end2end.yml \
+  ${EXTRA_ARGS}
diff --git a/experiments/scripts/rfcn_end2end.sh b/experiments/scripts/rfcn_end2end.sh
new file mode 100755
index 0000000..c54f9e1
--- /dev/null
+++ b/experiments/scripts/rfcn_end2end.sh
@@ -0,0 +1,68 @@
+#!/bin/bash
+# Usage:
+# ./experiments/scripts/rfcn_end2end.sh GPU NET DATASET [options args to {train,test}_net.py]
+# DATASET is either pascal_voc or coco.
+#
+# Example:
+# ./experiments/scripts/rfcn_end2end.sh 0 ResNet50 pascal_voc \
+#   --set EXP_DIR foobar RNG_SEED 42 TRAIN.SCALES "[400, 500, 600, 700]"
+
+set -x
+set -e
+
+export PYTHONUNBUFFERED="True"
+
+GPU_ID=$1
+NET=$2
+NET_lc=${NET,,}
+DATASET=$3
+
+array=( $@ )
+len=${#array[@]}
+EXTRA_ARGS=${array[@]:3:$len}
+EXTRA_ARGS_SLUG=${EXTRA_ARGS// /_}
+
+case $DATASET in
+  pascal_voc)
+    TRAIN_IMDB="voc_0712_trainval"
+    TEST_IMDB="voc_0712_test"
+    PT_DIR="pascal_voc"
+    ITERS=120000
+    ;;
+  coco)
+    # This is a very long and slow training schedule
+    # You can probably use fewer iterations and reduce the
+    # time to the LR drop (set in the solver to 350,000 iterations).
+    TRAIN_IMDB="coco_2014_train"
+    TEST_IMDB="coco_2014_minival"
+    PT_DIR="coco"
+    ITERS=490000
+    ;;
+  *)
+    echo "No dataset given"
+    exit
+    ;;
+esac
+
+LOG="experiments/logs/rfcn_end2end_${NET}_${EXTRA_ARGS_SLUG}.txt.`date +'%Y-%m-%d_%H-%M-%S'`"
+exec &> >(tee -a "$LOG")
+echo Logging output to "$LOG"
+
+time ./tools/train_net.py --gpu ${GPU_ID} \
+  --solver models/${PT_DIR}/${NET}/rfcn_end2end/solver.prototxt \
+  --weights data/imagenet_models/${NET}-model.caffemodel \
+  --imdb ${TRAIN_IMDB} \
+  --iters ${ITERS} \
+  --cfg experiments/cfgs/rfcn_end2end.yml \
+  ${EXTRA_ARGS}
+
+set +x
+NET_FINAL=`grep -B 1 "done solving" ${LOG} | grep "Wrote snapshot" | awk '{print $4}'`
+set -x
+
+time ./tools/test_net.py --gpu ${GPU_ID} \
+  --def models/${PT_DIR}/${NET}/rfcn_end2end/test_agonistic.prototxt \
+  --net ${NET_FINAL} \
+  --imdb ${TEST_IMDB} \
+  --cfg experiments/cfgs/rfcn_end2end.yml \
+  ${EXTRA_ARGS}
diff --git a/experiments/scripts/rfcn_end2end_ohem.sh b/experiments/scripts/rfcn_end2end_ohem.sh
new file mode 100755
index 0000000..ed42477
--- /dev/null
+++ b/experiments/scripts/rfcn_end2end_ohem.sh
@@ -0,0 +1,70 @@
+#!/bin/bash
+# Usage:
+# ./experiments/scripts/rfcn_end2end_ohem.sh GPU NET DATASET [options args to {train,test}_net.py]
+# DATASET is either pascal_voc or coco.
+#
+# Example:
+# ./experiments/scripts/rfcn_end2end_ohem.sh 0 ResNet50 pascal_voc \
+#   --set EXP_DIR foobar RNG_SEED 42 TRAIN.SCALES "[400, 500, 600, 700]"
+
+set -x
+set -e
+
+export PYTHONUNBUFFERED="True"
+
+GPU_ID=$1
+NET=$2
+NET_lc=${NET,,}
+DATASET=$3
+
+array=( $@ )
+len=${#array[@]}
+EXTRA_ARGS=${array[@]:3:$len}
+EXTRA_ARGS_SLUG=${EXTRA_ARGS// /_}
+
+case $DATASET in
+  pascal_voc)
+    TRAIN_IMDB="voc_0712_trainval"
+    TEST_IMDB="voc_0712_test"
+    PT_DIR="pascal_voc"
+    ITERS=120000
+    ;;
+  coco)
+    # This is a very long and slow training schedule
+    # You can probably use fewer iterations and reduce the
+    # time to the LR drop (set in the solver to 350,000 iterations).
+    TRAIN_IMDB="coco_2014_train"
+    TEST_IMDB="coco_2014_minival"
+    PT_DIR="coco"
+    ITERS=490000
+    ;;
+  *)
+    echo "No dataset given"
+    exit
+    ;;
+esac
+
+LOG="experiments/logs/rfcn_end2end_${NET}_${EXTRA_ARGS_SLUG}.txt.`date +'%Y-%m-%d_%H-%M-%S'`"
+exec &> >(tee -a "$LOG")
+echo Logging output to "$LOG"
+
+
+time ./tools/train_net.py --gpu ${GPU_ID} \
+  --solver models/${PT_DIR}/${NET}/rfcn_end2end/solver_ohem.prototxt \
+  --weights data/imagenet_models/${NET}-model.caffemodel \
+  --imdb ${TRAIN_IMDB} \
+  --iters ${ITERS} \
+  --cfg experiments/cfgs/rfcn_end2end_ohem.yml \
+  ${EXTRA_ARGS}
+
+
+set +x
+NET_FINAL=`tail -n 100 ${LOG} | grep -B 1 "done solving" | grep "Wrote snapshot" | awk '{print $4}'`
+set -x
+
+time ./tools/test_net.py --gpu ${GPU_ID} \
+  --def models/${PT_DIR}/${NET}/rfcn_end2end/test_agonistic.prototxt \
+  --net ${NET_FINAL} \
+  --imdb ${TEST_IMDB} \
+  --cfg experiments/cfgs/rfcn_end2end_ohem.yml \
+  ${EXTRA_ARGS}
diff --git a/experiments/scripts/rfcn_end2end_ohem_warmup.sh b/experiments/scripts/rfcn_end2end_ohem_warmup.sh
new file mode 100755
index 0000000..4efbac7
--- /dev/null
+++ b/experiments/scripts/rfcn_end2end_ohem_warmup.sh
@@ -0,0 +1,84 @@
+#!/bin/bash
+# Usage:
+# ./experiments/scripts/rfcn_end2end_ohem.sh GPU NET DATASET [options args to {train,test}_net.py]
+# DATASET is either pascal_voc or coco.
+#
+# Example:
+# ./experiments/scripts/rfcn_end2end_ohem.sh 0 ResNet50 pascal_voc \
+#   --set EXP_DIR foobar RNG_SEED 42 TRAIN.SCALES "[400, 500, 600, 700]"
+
+set -x
+set -e
+
+export PYTHONUNBUFFERED="True"
+
+GPU_ID=$1
+NET=$2
+NET_lc=${NET,,}
+DATASET=$3
+
+array=( $@ )
+len=${#array[@]}
+EXTRA_ARGS=${array[@]:3:$len}
+EXTRA_ARGS_SLUG=${EXTRA_ARGS// /_}
+
+case $DATASET in
+  pascal_voc)
+    TRAIN_IMDB="voc_0712_trainval"
+    TEST_IMDB="voc_0712_test"
+    PT_DIR="pascal_voc"
+    WARMUP_ITERS=10000
+    ITERS=150000
+    ;;
+  coco)
+    # This is a very long and slow training schedule
+    # You can probably use fewer iterations and reduce the
+    # time to the LR drop (set in the solver to 350,000 iterations).
+    TRAIN_IMDB="coco_2014_train"
+    TEST_IMDB="coco_2014_minival"
+    PT_DIR="coco"
+    WARMUP_ITERS=10000
+    ITERS=480000
+    ;;
+  *)
+    echo "No dataset given"
+    exit
+    ;;
+esac
+
+LOG="experiments/logs/rfcn_end2end_${NET}_${EXTRA_ARGS_SLUG}.txt.`date +'%Y-%m-%d_%H-%M-%S'`"
+exec &> >(tee -a "$LOG")
+echo Logging output to "$LOG"
+
+time ./tools/train_net.py --gpu ${GPU_ID} \
+  --solver models/${PT_DIR}/${NET}/rfcn_end2end/solver_warmup.prototxt \
+  --weights data/imagenet_models/${NET}-model.caffemodel \
+  --imdb ${TRAIN_IMDB} \
+  --iters ${WARMUP_ITERS} \
+  --cfg experiments/cfgs/rfcn_end2end_ohem.yml \
+  ${EXTRA_ARGS}
+
+set +x
+NET_CONTINUE=`tail -n 100 ${LOG} | grep -B 1 "done solving" | grep "Wrote snapshot" | awk '{print $4}'`
+set -x
+
+
+time ./tools/train_net.py --gpu ${GPU_ID} \
+  --solver models/${PT_DIR}/${NET}/rfcn_end2end/solver_continue.prototxt \
+  --weights ${NET_CONTINUE} \
+  --imdb ${TRAIN_IMDB} \
+  --iters ${ITERS} \
+  --cfg experiments/cfgs/rfcn_end2end_ohem.yml \
+  ${EXTRA_ARGS}
+
+
+set +x
+NET_FINAL=`tail -n 100 ${LOG} | grep -B 1 "done solving" | grep "Wrote snapshot" | awk '{print $4}'`
+set -x
+
+time ./tools/test_net.py --gpu ${GPU_ID} \
+  --def models/${PT_DIR}/${NET}/rfcn_end2end/test_agonistic.prototxt \
+  --net ${NET_FINAL} \
+  --imdb ${TEST_IMDB} \
+  --cfg experiments/cfgs/rfcn_end2end_ohem.yml \
+  ${EXTRA_ARGS}
diff --git a/lib/Makefile b/lib/Makefile
new file mode 100644
index 0000000..a482398
--- /dev/null
+++ b/lib/Makefile
@@ -0,0 +1,3 @@
+all:
+	python setup.py build_ext --inplace
+	rm -rf build
diff --git a/lib/datasets/VOCdevkit-matlab-wrapper/get_voc_opts.m b/lib/datasets/VOCdevkit-matlab-wrapper/get_voc_opts.m
new file mode 100644
index 0000000..629597a
--- /dev/null
+++ b/lib/datasets/VOCdevkit-matlab-wrapper/get_voc_opts.m
@@ -0,0 +1,14 @@
+function VOCopts = get_voc_opts(path)
+
+tmp = pwd;
+cd(path);
+try
+  addpath('VOCcode');
+  VOCinit;
+catch
+  rmpath('VOCcode');
+  cd(tmp);
+  error(sprintf('VOCcode directory not found under %s', path));
+end
+rmpath('VOCcode');
+cd(tmp);
diff --git a/lib/datasets/VOCdevkit-matlab-wrapper/voc_eval.m b/lib/datasets/VOCdevkit-matlab-wrapper/voc_eval.m
new file mode 100644
index 0000000..1911a0e
--- /dev/null
+++ b/lib/datasets/VOCdevkit-matlab-wrapper/voc_eval.m
@@ -0,0 +1,56 @@
+function res = voc_eval(path, comp_id, test_set, output_dir)
+
+VOCopts = get_voc_opts(path);
+VOCopts.testset = test_set;
+
+for i = 1:length(VOCopts.classes)
+  cls = VOCopts.classes{i};
+  res(i) = voc_eval_cls(cls, VOCopts, comp_id, output_dir);
+end
+
+fprintf('\n~~~~~~~~~~~~~~~~~~~~\n');
+fprintf('Results:\n');
+aps = [res(:).ap]';
+fprintf('%.1f\n', aps * 100);
+fprintf('%.1f\n', mean(aps) * 100);
+fprintf('~~~~~~~~~~~~~~~~~~~~\n');
+
+function res = voc_eval_cls(cls, VOCopts, comp_id, output_dir)
+
+test_set = VOCopts.testset;
+year = VOCopts.dataset(4:end);
+
+addpath(fullfile(VOCopts.datadir, 'VOCcode'));
+
+res_fn = sprintf(VOCopts.detrespath, comp_id, cls);
+
+recall = [];
+prec = [];
+ap = 0;
+ap_auc = 0;
+
+do_eval = (str2num(year) <= 2007) | ~strcmp(test_set, 'test');
+if do_eval
+  % Bug in VOCevaldet requires that tic has been called first
+  tic;
+  [recall, prec, ap] = VOCevaldet(VOCopts, comp_id, cls, true);
+  ap_auc = xVOCap(recall, prec);
+
+  % force plot limits
+  ylim([0 1]);
+  xlim([0 1]);
+
+  print(gcf, '-djpeg', '-r0', ...
+        [output_dir '/' cls '_pr.jpg']);
+end
+fprintf('!!! %s : %.4f %.4f\n', cls, ap, ap_auc);
+
+res.recall = recall;
+res.prec = prec;
+res.ap = ap;
+res.ap_auc = ap_auc;
+
+save([output_dir '/' cls '_pr.mat'], ...
+     'res', 'recall', 'prec', 'ap', 'ap_auc');
+
+rmpath(fullfile(VOCopts.datadir, 'VOCcode'));
diff --git a/lib/datasets/VOCdevkit-matlab-wrapper/xVOCap.m b/lib/datasets/VOCdevkit-matlab-wrapper/xVOCap.m
new file mode 100644
index 0000000..de6c628
--- /dev/null
+++ b/lib/datasets/VOCdevkit-matlab-wrapper/xVOCap.m
@@ -0,0 +1,10 @@
+function ap = xVOCap(rec,prec)
+% From the PASCAL VOC 2011 devkit
+
+mrec=[0 ; rec ; 1];
+mpre=[0 ; prec ; 0];
+for i=numel(mpre)-1:-1:1
+    mpre(i)=max(mpre(i),mpre(i+1));
+end
+i=find(mrec(2:end)~=mrec(1:end-1))+1;
+ap=sum((mrec(i)-mrec(i-1)).*mpre(i));
diff --git a/lib/datasets/__init__.py b/lib/datasets/__init__.py
new file mode 100644
index 0000000..7ba6a65
--- /dev/null
+++ b/lib/datasets/__init__.py
@@ -0,0 +1,6 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
diff --git a/lib/datasets/coco.py b/lib/datasets/coco.py
new file mode 100644
index 0000000..bfe8ff3
--- /dev/null
+++ b/lib/datasets/coco.py
@@ -0,0 +1,394 @@
+# --------------------------------------------------------
+# Fast/er R-CNN
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+from datasets.imdb import imdb
+import datasets.ds_utils as ds_utils
+from fast_rcnn.config import cfg
+import os.path as osp
+import sys
+import os
+import numpy as np
+import scipy.sparse
+import scipy.io as sio
+import cPickle
+import json
+import uuid
+# COCO API
+from pycocotools.coco import COCO
+from pycocotools.cocoeval import COCOeval
+from pycocotools import mask as COCOmask
+
+def _filter_crowd_proposals(roidb, crowd_thresh):
+    """
+    Finds proposals that are inside crowd regions and marks them with
+    overlap = -1 (for all gt rois), which means they will be excluded from
+    training.
+    """
+    for ix, entry in enumerate(roidb):
+        overlaps = entry['gt_overlaps'].toarray()
+        crowd_inds = np.where(overlaps.max(axis=1) == -1)[0]
+        non_gt_inds = np.where(entry['gt_classes'] == 0)[0]
+        if len(crowd_inds) == 0 or len(non_gt_inds) == 0:
+            continue
+        iscrowd = [int(True) for _ in xrange(len(crowd_inds))]
+        crowd_boxes = ds_utils.xyxy_to_xywh(entry['boxes'][crowd_inds, :])
+        non_gt_boxes = ds_utils.xyxy_to_xywh(entry['boxes'][non_gt_inds, :])
+        ious = COCOmask.iou(non_gt_boxes, crowd_boxes, iscrowd)
+        bad_inds = np.where(ious.max(axis=1) > crowd_thresh)[0]
+        overlaps[non_gt_inds[bad_inds], :] = -1
+        roidb[ix]['gt_overlaps'] = scipy.sparse.csr_matrix(overlaps)
+    return roidb
+
+class coco(imdb):
+    def __init__(self, image_set, year):
+        imdb.__init__(self, 'coco_' + year + '_' + image_set)
+        # COCO specific config options
+        self.config = {'top_k' : 2000,
+                       'use_salt' : True,
+                       'cleanup' : True,
+                       'crowd_thresh' : 0.7,
+                       'min_size' : 2}
+        # name, paths
+        self._year = year
+        self._image_set = image_set
+        self._data_path = osp.join(cfg.DATA_DIR, 'coco')
+        # load COCO API, classes, class <-> id mappings
+        self._COCO = COCO(self._get_ann_file())
+        cats = self._COCO.loadCats(self._COCO.getCatIds())
+        self._classes = tuple(['__background__'] + [c['name'] for c in cats])
+        self._class_to_ind = dict(zip(self.classes, xrange(self.num_classes)))
+        self._class_to_coco_cat_id = dict(zip([c['name'] for c in cats],
+                                              self._COCO.getCatIds()))
+        self._image_index = self._load_image_set_index()
+        # Default to roidb handler
+        self.set_proposal_method('selective_search')
+        self.competition_mode(False)
+
+        # Some image sets are "views" (i.e. subsets) into others.
+        # For example, minival2014 is a random 5000 image subset of val2014.
+        # This mapping tells us where the view's images and proposals come from.
+        self._view_map = {
+            'minival2014' : 'val2014',          # 5k val2014 subset
+            'valminusminival2014' : 'val2014',  # val2014 \setminus minival2014
+        }
+        coco_name = image_set + year  # e.g., "val2014"
+        self._data_name = (self._view_map[coco_name]
+                           if self._view_map.has_key(coco_name)
+                           else coco_name)
+        # Dataset splits that have ground-truth annotations (test splits
+        # do not have gt annotations)
+        self._gt_splits = ('train', 'val', 'minival')
+
+    def _get_ann_file(self):
+        prefix = 'instances' if self._image_set.find('test') == -1 \
+                             else 'image_info'
+        return osp.join(self._data_path, 'annotations',
+                        prefix + '_' + self._image_set + self._year + '.json')
+
+    def _load_image_set_index(self):
+        """
+        Load image ids.
+        """
+        image_ids = self._COCO.getImgIds()
+        return image_ids
+
+    def _get_widths(self):
+        anns = self._COCO.loadImgs(self._image_index)
+        widths = [ann['width'] for ann in anns]
+        return widths
+
+    def image_path_at(self, i):
+        """
+        Return the absolute path to image i in the image sequence.
+        """
+        return self.image_path_from_index(self._image_index[i])
+
+    def image_path_from_index(self, index):
+        """
+        Construct an image path from the image's "index" identifier.
+        """
+        # Example image path for index=119993:
+        #   images/train2014/COCO_train2014_000000119993.jpg
+        file_name = ('COCO_' + self._data_name + '_' +
+                     str(index).zfill(12) + '.jpg')
+        image_path = osp.join(self._data_path, 'images',
+                              self._data_name, file_name)
+        assert osp.exists(image_path), \
+                'Path does not exist: {}'.format(image_path)
+        return image_path
+
+    def selective_search_roidb(self):
+        return self._roidb_from_proposals('selective_search')
+
+    def edge_boxes_roidb(self):
+        return self._roidb_from_proposals('edge_boxes_AR')
+
+    def mcg_roidb(self):
+        return self._roidb_from_proposals('MCG')
+
+    def _roidb_from_proposals(self, method):
+        """
+        Creates a roidb from pre-computed proposals of a particular methods.
+        """
+        top_k = self.config['top_k']
+        cache_file = osp.join(self.cache_path, self.name +
+                              '_{:s}_top{:d}'.format(method, top_k) +
+                              '_roidb.pkl')
+
+        if osp.exists(cache_file):
+            with open(cache_file, 'rb') as fid:
+                roidb = cPickle.load(fid)
+            print '{:s} {:s} roidb loaded from {:s}'.format(self.name, method,
+                                                            cache_file)
+            return roidb
+
+        if self._image_set in self._gt_splits:
+            gt_roidb = self.gt_roidb()
+            method_roidb = self._load_proposals(method, gt_roidb)
+            roidb = imdb.merge_roidbs(gt_roidb, method_roidb)
+            # Make sure we don't use proposals that are contained in crowds
+            roidb = _filter_crowd_proposals(roidb, self.config['crowd_thresh'])
+        else:
+            roidb = self._load_proposals(method, None)
+        with open(cache_file, 'wb') as fid:
+            cPickle.dump(roidb, fid, cPickle.HIGHEST_PROTOCOL)
+        print 'wrote {:s} roidb to {:s}'.format(method, cache_file)
+        return roidb
+
+    def _load_proposals(self, method, gt_roidb):
+        """
+        Load pre-computed proposals in the format provided by Jan Hosang:
+        http://www.mpi-inf.mpg.de/departments/computer-vision-and-multimodal-
+          computing/research/object-recognition-and-scene-understanding/how-
+          good-are-detection-proposals-really/
+        For MCG, use boxes from http://www.eecs.berkeley.edu/Research/Projects/
+          CS/vision/grouping/mcg/ and convert the file layout using
+        lib/datasets/tools/mcg_munge.py.
+        """
+        box_list = []
+        top_k = self.config['top_k']
+        valid_methods = [
+            'MCG',
+            'selective_search',
+            'edge_boxes_AR',
+            'edge_boxes_70']
+        assert method in valid_methods
+
+        print 'Loading {} boxes'.format(method)
+        for i, index in enumerate(self._image_index):
+            if i % 1000 == 0:
+                print '{:d} / {:d}'.format(i + 1, len(self._image_index))
+
+            box_file = osp.join(
+                cfg.DATA_DIR, 'coco_proposals', method, 'mat',
+                self._get_box_file(index))
+
+            raw_data = sio.loadmat(box_file)['boxes']
+            boxes = np.maximum(raw_data - 1, 0).astype(np.uint16)
+            if method == 'MCG':
+                # Boxes from the MCG website are in (y1, x1, y2, x2) order
+                boxes = boxes[:, (1, 0, 3, 2)]
+            # Remove duplicate boxes and very small boxes and then take top k
+            keep = ds_utils.unique_boxes(boxes)
+            boxes = boxes[keep, :]
+            keep = ds_utils.filter_small_boxes(boxes, self.config['min_size'])
+            boxes = boxes[keep, :]
+            boxes = boxes[:top_k, :]
+            box_list.append(boxes)
+            # Sanity check
+            im_ann = self._COCO.loadImgs(index)[0]
+            width = im_ann['width']
+            height = im_ann['height']
+            ds_utils.validate_boxes(boxes, width=width, height=height)
+        return self.create_roidb_from_box_list(box_list, gt_roidb)
+
+    def gt_roidb(self):
+        """
+        Return the database of ground-truth regions of interest.
+        This function loads/saves from/to a cache file to speed up future calls.
+        """
+        cache_file = osp.join(self.cache_path, self.name + '_gt_roidb.pkl')
+        if osp.exists(cache_file):
+            with open(cache_file, 'rb') as fid:
+                roidb = cPickle.load(fid)
+            print '{} gt roidb loaded from {}'.format(self.name, cache_file)
+            return roidb
+
+        gt_roidb = [self._load_coco_annotation(index)
+                    for index in self._image_index]
+
+        with open(cache_file, 'wb') as fid:
+            cPickle.dump(gt_roidb, fid, cPickle.HIGHEST_PROTOCOL)
+        print 'wrote gt roidb to {}'.format(cache_file)
+        return gt_roidb
+
+    def _load_coco_annotation(self, index):
+        """
+        Loads COCO bounding-box instance annotations. Crowd instances are
+        handled by marking their overlaps (with all categories) to -1. This
+        overlap value means that crowd "instances" are excluded from training.
+        """
+        im_ann = self._COCO.loadImgs(index)[0]
+        width = im_ann['width']
+        height = im_ann['height']
+
+        annIds = self._COCO.getAnnIds(imgIds=index, iscrowd=None)
+        objs = self._COCO.loadAnns(annIds)
+        # Sanitize bboxes -- some are invalid
+        valid_objs = []
+        for obj in objs:
+            x1 = np.max((0, obj['bbox'][0]))
+            y1 = np.max((0, obj['bbox'][1]))
+            x2 = np.min((width - 1, x1 + np.max((0, obj['bbox'][2] - 1))))
+            y2 = np.min((height - 1, y1 + np.max((0, obj['bbox'][3] - 1))))
+            if obj['area'] > 0 and x2 >= x1 and y2 >= y1:
+                obj['clean_bbox'] = [x1, y1, x2, y2]
+                valid_objs.append(obj)
+        objs = valid_objs
+        num_objs = len(objs)
+
+        boxes = np.zeros((num_objs, 4), dtype=np.uint16)
+        gt_classes = np.zeros((num_objs), dtype=np.int32)
+        overlaps = np.zeros((num_objs, self.num_classes), dtype=np.float32)
+        seg_areas = np.zeros((num_objs), dtype=np.float32)
+
+        # Lookup table to map from COCO category ids to our internal class
+        # indices
+        coco_cat_id_to_class_ind = dict([(self._class_to_coco_cat_id[cls],
+                                          self._class_to_ind[cls])
+                                         for cls in self._classes[1:]])
+
+        for ix, obj in enumerate(objs):
+            cls = coco_cat_id_to_class_ind[obj['category_id']]
+            boxes[ix, :] = obj['clean_bbox']
+            gt_classes[ix] = cls
+            seg_areas[ix] = obj['area']
+            if obj['iscrowd']:
+                # Set overlap to -1 for all classes for crowd objects
+                # so they will be excluded during training
+                overlaps[ix, :] = -1.0
+            else:
+                overlaps[ix, cls] = 1.0
+
+        ds_utils.validate_boxes(boxes, width=width, height=height)
+        overlaps = scipy.sparse.csr_matrix(overlaps)
+        return {'boxes' : boxes,
+                'gt_classes': gt_classes,
+                'gt_overlaps' : overlaps,
+                'flipped' : False,
+                'seg_areas' : seg_areas}
+
+    def _get_box_file(self, index):
+        # first 14 chars / first 22 chars / all chars + .mat
+        # COCO_val2014_0/COCO_val2014_000000447/COCO_val2014_000000447991.mat
+        file_name = ('COCO_' + self._data_name +
+                     '_' + str(index).zfill(12) + '.mat')
+        return osp.join(file_name[:14], file_name[:22], file_name)
+
+    def _print_detection_eval_metrics(self, coco_eval):
+        IoU_lo_thresh = 0.5
+        IoU_hi_thresh = 0.95
+        def _get_thr_ind(coco_eval, thr):
+            ind = np.where((coco_eval.params.iouThrs > thr - 1e-5) &
+                           (coco_eval.params.iouThrs < thr + 1e-5))[0][0]
+            iou_thr = coco_eval.params.iouThrs[ind]
+            assert np.isclose(iou_thr, thr)
+            return ind
+
+        ind_lo = _get_thr_ind(coco_eval, IoU_lo_thresh)
+        ind_hi = _get_thr_ind(coco_eval, IoU_hi_thresh)
+        # precision has dims (iou, recall, cls, area range, max dets)
+        # area range index 0: all area ranges
+        # max dets index 2: 100 per image
+        precision = \
+            coco_eval.eval['precision'][ind_lo:(ind_hi + 1), :, :, 0, 2]
+        ap_default = np.mean(precision[precision > -1])
+        print ('~~~~ Mean and per-category AP @ IoU=[{:.2f},{:.2f}] '
+               '~~~~').format(IoU_lo_thresh, IoU_hi_thresh)
+        print '{:.1f}'.format(100 * ap_default)
+        for cls_ind, cls in enumerate(self.classes):
+            if cls == '__background__':
+                continue
+            # minus 1 because of __background__
+            precision = coco_eval.eval['precision'][ind_lo:(ind_hi + 1), :, cls_ind - 1, 0, 2]
+            ap = np.mean(precision[precision > -1])
+            print '{:.1f}'.format(100 * ap)
+
+        print '~~~~ Summary metrics ~~~~'
+        coco_eval.summarize()
+
+    def _do_detection_eval(self, res_file, output_dir):
+        ann_type = 'bbox'
+        coco_dt = self._COCO.loadRes(res_file)
+        coco_eval = COCOeval(self._COCO, coco_dt)
+        coco_eval.params.useSegm = (ann_type == 'segm')
+        coco_eval.evaluate()
+        coco_eval.accumulate()
+        self._print_detection_eval_metrics(coco_eval)
+        eval_file = osp.join(output_dir, 'detection_results.pkl')
+        with open(eval_file, 'wb') as fid:
+            cPickle.dump(coco_eval, fid, cPickle.HIGHEST_PROTOCOL)
+        print 'Wrote COCO eval results to: {}'.format(eval_file)
+
+    def _coco_results_one_category(self, boxes, cat_id):
+        results = []
+        for im_ind, index in enumerate(self.image_index):
+            dets = boxes[im_ind].astype(np.float)
+            if dets == []:
+                continue
+            scores = dets[:, -1]
+            xs = dets[:, 0]
+            ys = dets[:, 1]
+            ws = dets[:, 2] - xs + 1
+            hs = dets[:, 3] - ys + 1
+            results.extend(
+              [{'image_id' : index,
+                'category_id' : cat_id,
+                'bbox' : [xs[k], ys[k], ws[k], hs[k]],
+                'score' : scores[k]} for k in xrange(dets.shape[0])])
+        return results
+
+    def _write_coco_results_file(self, all_boxes, res_file):
+        # [{"image_id": 42,
+        #   "category_id": 18,
+        #   "bbox": [258.15,41.29,348.26,243.78],
+        #   "score": 0.236}, ...]
+        results = []
+        for cls_ind, cls in enumerate(self.classes):
+            if cls == '__background__':
+                continue
+            print 'Collecting {} results ({:d}/{:d})'.format(cls, cls_ind,
+                                                          self.num_classes - 1)
+            coco_cat_id = self._class_to_coco_cat_id[cls]
+            results.extend(self._coco_results_one_category(all_boxes[cls_ind],
+                                                           coco_cat_id))
+        print 'Writing results json to {}'.format(res_file)
+        with open(res_file, 'w') as fid:
+            json.dump(results, fid)
+
+    def evaluate_detections(self, all_boxes, output_dir):
+        res_file = osp.join(output_dir, ('detections_' +
+                                         self._image_set +
+                                         self._year +
+                                         '_results'))
+        if self.config['use_salt']:
+            res_file += '_{}'.format(str(uuid.uuid4()))
+        res_file += '.json'
+        self._write_coco_results_file(all_boxes, res_file)
+        # Only do evaluation on non-test sets
+        if self._image_set.find('test') == -1:
+            self._do_detection_eval(res_file, output_dir)
+        # Optionally cleanup results json file
+        if self.config['cleanup']:
+            os.remove(res_file)
+
+    def competition_mode(self, on):
+        if on:
+            self.config['use_salt'] = False
+            self.config['cleanup'] = False
+        else:
+            self.config['use_salt'] = True
+            self.config['cleanup'] = True
diff --git a/lib/datasets/ds_utils.py b/lib/datasets/ds_utils.py
new file mode 100644
index 0000000..f66a7f6
--- /dev/null
+++ b/lib/datasets/ds_utils.py
@@ -0,0 +1,41 @@
+# --------------------------------------------------------
+# Fast/er R-CNN
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+import numpy as np
+
+def unique_boxes(boxes, scale=1.0):
+    """Return indices of unique boxes."""
+    v = np.array([1, 1e3, 1e6, 1e9])
+    hashes = np.round(boxes * scale).dot(v)
+    _, index = np.unique(hashes, return_index=True)
+    return np.sort(index)
+
+def xywh_to_xyxy(boxes):
+    """Convert [x y w h] box format to [x1 y1 x2 y2] format."""
+    return np.hstack((boxes[:, 0:2], boxes[:, 0:2] + boxes[:, 2:4] - 1))
+
+def xyxy_to_xywh(boxes):
+    """Convert [x1 y1 x2 y2] box format to [x y w h] format."""
+    return np.hstack((boxes[:, 0:2], boxes[:, 2:4] - boxes[:, 0:2] + 1))
+
+def validate_boxes(boxes, width=0, height=0):
+    """Check that a set of boxes are valid."""
+    x1 = boxes[:, 0]
+    y1 = boxes[:, 1]
+    x2 = boxes[:, 2]
+    y2 = boxes[:, 3]
+    assert (x1 >= 0).all()
+    assert (y1 >= 0).all()
+    assert (x2 >= x1).all()
+    assert (y2 >= y1).all()
+    assert (x2 < width).all()
+    assert (y2 < height).all()
+
+def filter_small_boxes(boxes, min_size):
+    w = boxes[:, 2] - boxes[:, 0]
+    h = boxes[:, 3] - boxes[:, 1]
+    keep = np.where((w >= min_size) & (h > min_size))[0]
+    return keep
diff --git a/lib/datasets/factory.py b/lib/datasets/factory.py
new file mode 100644
index 0000000..339ea13
--- /dev/null
+++ b/lib/datasets/factory.py
@@ -0,0 +1,43 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Factory method for easily getting imdbs by name."""
+
+__sets = {}
+
+from datasets.pascal_voc import pascal_voc
+from datasets.coco import coco
+import numpy as np
+
+# Set up voc_<year>_<split> using selective search "fast" mode
+for year in ['2007', '2012', '0712']:
+    for split in ['train', 'val', 'trainval', 'test']:
+        name = 'voc_{}_{}'.format(year, split)
+        __sets[name] = (lambda split=split, year=year: pascal_voc(split, year))
+
+
+# Set up coco_2014_<split>
+for year in ['2014']:
+    for split in ['train', 'val', 'minival', 'valminusminival']:
+        name = 'coco_{}_{}'.format(year, split)
+        __sets[name] = (lambda split=split, year=year: coco(split, year))
+
+# Set up coco_2015_<split>
+for year in ['2015']:
+    for split in ['test', 'test-dev']:
+        name = 'coco_{}_{}'.format(year, split)
+        __sets[name] = (lambda split=split, year=year: coco(split, year))
+
+def get_imdb(name):
+    """Get an imdb (image database) by name."""
+    if not __sets.has_key(name):
+        raise KeyError('Unknown dataset: {}'.format(name))
+    return __sets[name]()
+
+def list_imdbs():
+    """List all registered imdbs."""
+    return __sets.keys()
diff --git a/lib/datasets/imdb.py b/lib/datasets/imdb.py
new file mode 100644
index 0000000..b56bf0a
--- /dev/null
+++ b/lib/datasets/imdb.py
@@ -0,0 +1,253 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+import os
+import os.path as osp
+import PIL
+from utils.cython_bbox import bbox_overlaps
+import numpy as np
+import scipy.sparse
+from fast_rcnn.config import cfg
+
+class imdb(object):
+    """Image database."""
+
+    def __init__(self, name):
+        self._name = name
+        self._num_classes = 0
+        self._classes = []
+        self._image_index = []
+        self._obj_proposer = 'selective_search'
+        self._roidb = None
+        self._roidb_handler = self.default_roidb
+        # Use this dict for storing dataset specific config options
+        self.config = {}
+
+    @property
+    def name(self):
+        return self._name
+
+    @property
+    def num_classes(self):
+        return len(self._classes)
+
+    @property
+    def classes(self):
+        return self._classes
+
+    @property
+    def image_index(self):
+        return self._image_index
+
+    @property
+    def roidb_handler(self):
+        return self._roidb_handler
+
+    @roidb_handler.setter
+    def roidb_handler(self, val):
+        self._roidb_handler = val
+
+    def set_proposal_method(self, method):
+        method = eval('self.' + method + '_roidb')
+        self.roidb_handler = method
+
+    @property
+    def roidb(self):
+        # A roidb is a list of dictionaries, each with the following keys:
+        #   boxes
+        #   gt_overlaps
+        #   gt_classes
+        #   flipped
+        if self._roidb is not None:
+            return self._roidb
+        self._roidb = self.roidb_handler()
+        return self._roidb
+
+    @property
+    def cache_path(self):
+        cache_path = osp.abspath(osp.join(cfg.DATA_DIR, 'cache'))
+        if not os.path.exists(cache_path):
+            os.makedirs(cache_path)
+        return cache_path
+
+    @property
+    def num_images(self):
+      return len(self.image_index)
+
+    def image_path_at(self, i):
+        raise NotImplementedError
+
+    def default_roidb(self):
+        raise NotImplementedError
+
+    def evaluate_detections(self, all_boxes, output_dir=None):
+        """
+        all_boxes is a list of length number-of-classes.
+        Each list element is a list of length number-of-images.
+        Each of those list elements is either an empty list []
+        or a numpy array of detection.
+
+        all_boxes[class][image] = [] or np.array of shape #dets x 5
+        """
+        raise NotImplementedError
+
+    def _get_widths(self):
+      return [PIL.Image.open(self.image_path_at(i)).size[0]
+              for i in xrange(self.num_images)]
+
+    def append_flipped_images(self):
+        num_images = self.num_images
+        widths = self._get_widths()
+        for i in xrange(num_images):
+            boxes = self.roidb[i]['boxes'].copy()
+            oldx1 = boxes[:, 0].copy()
+            oldx2 = boxes[:, 2].copy()
+            boxes[:, 0] = widths[i] - oldx2 - 1
+            boxes[:, 2] = widths[i] - oldx1 - 1
+            assert (boxes[:, 2] >= boxes[:, 0]).all()
+            entry = {'boxes' : boxes,
+                     'gt_overlaps' : self.roidb[i]['gt_overlaps'],
+                     'gt_classes' : self.roidb[i]['gt_classes'],
+                     'flipped' : True}
+            self.roidb.append(entry)
+        self._image_index = self._image_index * 2
+
+    def evaluate_recall(self, candidate_boxes=None, thresholds=None,
+                        area='all', limit=None):
+        """Evaluate detection proposal recall metrics.
+
+        Returns:
+            results: dictionary of results with keys
+                'ar': average recall
+                'recalls': vector recalls at each IoU overlap threshold
+                'thresholds': vector of IoU overlap thresholds
+                'gt_overlaps': vector of all ground-truth overlaps
+        """
+        # Record max overlap value for each gt box
+        # Return vector of overlap values
+        areas = { 'all': 0, 'small': 1, 'medium': 2, 'large': 3,
+                  '96-128': 4, '128-256': 5, '256-512': 6, '512-inf': 7}
+        area_ranges = [ [0**2, 1e5**2],    # all
+                        [0**2, 32**2],     # small
+                        [32**2, 96**2],    # medium
+                        [96**2, 1e5**2],   # large
+                        [96**2, 128**2],   # 96-128
+                        [128**2, 256**2],  # 128-256
+                        [256**2, 512**2],  # 256-512
+                        [512**2, 1e5**2],  # 512-inf
+                      ]
+        assert areas.has_key(area), 'unknown area range: {}'.format(area)
+        area_range = area_ranges[areas[area]]
+        gt_overlaps = np.zeros(0)
+        num_pos = 0
+        for i in xrange(self.num_images):
+            # Checking for max_overlaps == 1 avoids including crowd annotations
+            # (...pretty hacking :/)
+            max_gt_overlaps = self.roidb[i]['gt_overlaps'].toarray().max(axis=1)
+            gt_inds = np.where((self.roidb[i]['gt_classes'] > 0) &
+                               (max_gt_overlaps == 1))[0]
+            gt_boxes = self.roidb[i]['boxes'][gt_inds, :]
+            gt_areas = self.roidb[i]['seg_areas'][gt_inds]
+            valid_gt_inds = np.where((gt_areas >= area_range[0]) &
+                                     (gt_areas <= area_range[1]))[0]
+            gt_boxes = gt_boxes[valid_gt_inds, :]
+            num_pos += len(valid_gt_inds)
+
+            if candidate_boxes is None:
+                # If candidate_boxes is not supplied, the default is to use the
+                # non-ground-truth boxes from this roidb
+                non_gt_inds = np.where(self.roidb[i]['gt_classes'] == 0)[0]
+                boxes = self.roidb[i]['boxes'][non_gt_inds, :]
+            else:
+                boxes = candidate_boxes[i]
+            if boxes.shape[0] == 0:
+                continue
+            if limit is not None and boxes.shape[0] > limit:
+                boxes = boxes[:limit, :]
+
+            overlaps = bbox_overlaps(boxes.astype(np.float),
+                                     gt_boxes.astype(np.float))
+
+            _gt_overlaps = np.zeros((gt_boxes.shape[0]))
+            for j in xrange(gt_boxes.shape[0]):
+                # find which proposal box maximally covers each gt box
+                argmax_overlaps = overlaps.argmax(axis=0)
+                # and get the iou amount of coverage for each gt box
+                max_overlaps = overlaps.max(axis=0)
+                # find which gt box is 'best' covered (i.e. 'best' = most iou)
+                gt_ind = max_overlaps.argmax()
+                gt_ovr = max_overlaps.max()
+                assert(gt_ovr >= 0)
+                # find the proposal box that covers the best covered gt box
+                box_ind = argmax_overlaps[gt_ind]
+                # record the iou coverage of this gt box
+                _gt_overlaps[j] = overlaps[box_ind, gt_ind]
+                assert(_gt_overlaps[j] == gt_ovr)
+                # mark the proposal box and the gt box as used
+                overlaps[box_ind, :] = -1
+                overlaps[:, gt_ind] = -1
+            # append recorded iou coverage level
+            gt_overlaps = np.hstack((gt_overlaps, _gt_overlaps))
+
+        gt_overlaps = np.sort(gt_overlaps)
+        if thresholds is None:
+            step = 0.05
+            thresholds = np.arange(0.5, 0.95 + 1e-5, step)
+        recalls = np.zeros_like(thresholds)
+        # compute recall for each iou threshold
+        for i, t in enumerate(thresholds):
+            recalls[i] = (gt_overlaps >= t).sum() / float(num_pos)
+        # ar = 2 * np.trapz(recalls, thresholds)
+        ar = recalls.mean()
+        return {'ar': ar, 'recalls': recalls, 'thresholds': thresholds,
+                'gt_overlaps': gt_overlaps}
+
+    def create_roidb_from_box_list(self, box_list, gt_roidb):
+        assert len(box_list) == self.num_images, \
+                'Number of boxes must match number of ground-truth images'
+        roidb = []
+        for i in xrange(self.num_images):
+            boxes = box_list[i]
+            num_boxes = boxes.shape[0]
+            overlaps = np.zeros((num_boxes, self.num_classes), dtype=np.float32)
+
+            if gt_roidb is not None and gt_roidb[i]['boxes'].size > 0:
+                gt_boxes = gt_roidb[i]['boxes']
+                gt_classes = gt_roidb[i]['gt_classes']
+                gt_overlaps = bbox_overlaps(boxes.astype(np.float),
+                                            gt_boxes.astype(np.float))
+                argmaxes = gt_overlaps.argmax(axis=1)
+                maxes = gt_overlaps.max(axis=1)
+                I = np.where(maxes > 0)[0]
+                overlaps[I, gt_classes[argmaxes[I]]] = maxes[I]
+
+            overlaps = scipy.sparse.csr_matrix(overlaps)
+            roidb.append({
+                'boxes' : boxes,
+                'gt_classes' : np.zeros((num_boxes,), dtype=np.int32),
+                'gt_overlaps' : overlaps,
+                'flipped' : False,
+                'seg_areas' : np.zeros((num_boxes,), dtype=np.float32),
+            })
+        return roidb
+
+    @staticmethod
+    def merge_roidbs(a, b):
+        assert len(a) == len(b)
+        for i in xrange(len(a)):
+            a[i]['boxes'] = np.vstack((a[i]['boxes'], b[i]['boxes']))
+            a[i]['gt_classes'] = np.hstack((a[i]['gt_classes'],
+                                            b[i]['gt_classes']))
+            a[i]['gt_overlaps'] = scipy.sparse.vstack([a[i]['gt_overlaps'],
+                                                       b[i]['gt_overlaps']])
+            a[i]['seg_areas'] = np.hstack((a[i]['seg_areas'],
+                                           b[i]['seg_areas']))
+        return a
+
+    def competition_mode(self, on):
+        """Turn competition mode on or off."""
+        pass
diff --git a/lib/datasets/pascal_voc.py b/lib/datasets/pascal_voc.py
new file mode 100644
index 0000000..b55f2f6
--- /dev/null
+++ b/lib/datasets/pascal_voc.py
@@ -0,0 +1,344 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+import os
+from datasets.imdb import imdb
+import datasets.ds_utils as ds_utils
+import xml.etree.ElementTree as ET
+import numpy as np
+import scipy.sparse
+import scipy.io as sio
+import utils.cython_bbox
+import cPickle
+import subprocess
+import uuid
+from voc_eval import voc_eval
+from fast_rcnn.config import cfg
+
+class pascal_voc(imdb):
+    def __init__(self, image_set, year, devkit_path=None):
+        imdb.__init__(self, 'voc_' + year + '_' + image_set)
+        self._year = year
+        self._image_set = image_set
+        self._devkit_path = self._get_default_path() if devkit_path is None \
+                            else devkit_path
+        self._data_path = os.path.join(self._devkit_path, 'VOC' + self._year)
+        self._classes = ('__background__', # always index 0
+                         'aeroplane', 'bicycle', 'bird', 'boat',
+                         'bottle', 'bus', 'car', 'cat', 'chair',
+                         'cow', 'diningtable', 'dog', 'horse',
+                         'motorbike', 'person', 'pottedplant',
+                         'sheep', 'sofa', 'train', 'tvmonitor')
+        self._class_to_ind = dict(zip(self.classes, xrange(self.num_classes)))
+        self._image_ext = '.jpg'
+        self._image_index = self._load_image_set_index()
+        # Default to roidb handler
+        self._roidb_handler = self.selective_search_roidb
+        self._salt = str(uuid.uuid4())
+        self._comp_id = 'comp4'
+
+        # PASCAL specific config options
+        self.config = {'cleanup'     : True,
+                       'use_salt'    : True,
+                       'use_diff'    : False,
+                       'matlab_eval' : False,
+                       'rpn_file'    : None,
+                       'min_size'    : 2}
+
+        assert os.path.exists(self._devkit_path), \
+                'VOCdevkit path does not exist: {}'.format(self._devkit_path)
+        assert os.path.exists(self._data_path), \
+                'Path does not exist: {}'.format(self._data_path)
+
+    def image_path_at(self, i):
+        """
+        Return the absolute path to image i in the image sequence.
+        """
+        return self.image_path_from_index(self._image_index[i])
+
+    def image_path_from_index(self, index):
+        """
+        Construct an image path from the image's "index" identifier.
+        """
+        image_path = os.path.join(self._data_path, 'JPEGImages',
+                                  index + self._image_ext)
+        assert os.path.exists(image_path), \
+                'Path does not exist: {}'.format(image_path)
+        return image_path
+
+    def _load_image_set_index(self):
+        """
+        Load the indexes listed in this dataset's image set file.
+        """
+        # Example path to image set file:
+        # self._devkit_path + /VOCdevkit2007/VOC2007/ImageSets/Main/val.txt
+        image_set_file = os.path.join(self._data_path, 'ImageSets', 'Main',
+                                      self._image_set + '.txt')
+        assert os.path.exists(image_set_file), \
+                'Path does not exist: {}'.format(image_set_file)
+        with open(image_set_file) as f:
+            image_index = [x.strip() for x in f.readlines()]
+        return image_index
+
+    def _get_default_path(self):
+        """
+        Return the default path where PASCAL VOC is expected to be installed.
+        """
+        return os.path.join(cfg.DATA_DIR, 'VOCdevkit' + self._year)
+
+    def gt_roidb(self):
+        """
+        Return the database of ground-truth regions of interest.
+
+        This function loads/saves from/to a cache file to speed up future calls.
+        """
+        cache_file = os.path.join(self.cache_path, self.name + '_gt_roidb.pkl')
+        if os.path.exists(cache_file):
+            with open(cache_file, 'rb') as fid:
+                roidb = cPickle.load(fid)
+            print '{} gt roidb loaded from {}'.format(self.name, cache_file)
+            return roidb
+
+        gt_roidb = [self._load_pascal_annotation(index)
+                    for index in self.image_index]
+        with open(cache_file, 'wb') as fid:
+            cPickle.dump(gt_roidb, fid, cPickle.HIGHEST_PROTOCOL)
+        print 'wrote gt roidb to {}'.format(cache_file)
+
+        return gt_roidb
+
+    def selective_search_roidb(self):
+        """
+        Return the database of selective search regions of interest.
+        Ground-truth ROIs are also included.
+
+        This function loads/saves from/to a cache file to speed up future calls.
+        """
+        cache_file = os.path.join(self.cache_path,
+                                  self.name + '_selective_search_roidb.pkl')
+
+        if os.path.exists(cache_file):
+            with open(cache_file, 'rb') as fid:
+                roidb = cPickle.load(fid)
+            print '{} ss roidb loaded from {}'.format(self.name, cache_file)
+            return roidb
+
+        if int(self._year) == 2007 or self._image_set != 'test':
+            gt_roidb = self.gt_roidb()
+            ss_roidb = self._load_selective_search_roidb(gt_roidb)
+            roidb = imdb.merge_roidbs(gt_roidb, ss_roidb)
+        else:
+            roidb = self._load_selective_search_roidb(None)
+        with open(cache_file, 'wb') as fid:
+            cPickle.dump(roidb, fid, cPickle.HIGHEST_PROTOCOL)
+        print 'wrote ss roidb to {}'.format(cache_file)
+
+        return roidb
+
+    def rpn_roidb(self):
+        if int(self._year) == 2007 or self._image_set != 'test':
+            gt_roidb = self.gt_roidb()
+            rpn_roidb = self._load_rpn_roidb(gt_roidb)
+            roidb = imdb.merge_roidbs(gt_roidb, rpn_roidb)
+        else:
+            roidb = self._load_rpn_roidb(None)
+
+        return roidb
+
+    def _load_rpn_roidb(self, gt_roidb):
+        filename = self.config['rpn_file']
+        print 'loading {}'.format(filename)
+        assert os.path.exists(filename), \
+               'rpn data not found at: {}'.format(filename)
+        with open(filename, 'rb') as f:
+            box_list = cPickle.load(f)
+        return self.create_roidb_from_box_list(box_list, gt_roidb)
+
+    def _load_selective_search_roidb(self, gt_roidb):
+        filename = os.path.abspath(os.path.join(cfg.DATA_DIR,
+                                                'selective_search_data',
+                                                self.name + '.mat'))
+        assert os.path.exists(filename), \
+               'Selective search data not found at: {}'.format(filename)
+        raw_data = sio.loadmat(filename)['boxes'].ravel()
+
+        box_list = []
+        for i in xrange(raw_data.shape[0]):
+            boxes = raw_data[i][:, (1, 0, 3, 2)] - 1
+            keep = ds_utils.unique_boxes(boxes)
+            boxes = boxes[keep, :]
+            keep = ds_utils.filter_small_boxes(boxes, self.config['min_size'])
+            boxes = boxes[keep, :]
+            box_list.append(boxes)
+
+        return self.create_roidb_from_box_list(box_list, gt_roidb)
+
+    def _load_pascal_annotation(self, index):
+        """
+        Load image and bounding boxes info from XML file in the PASCAL VOC
+        format.
+        """
+        filename = os.path.join(self._data_path, 'Annotations', index + '.xml')
+        tree = ET.parse(filename)
+        objs = tree.findall('object')
+        if not self.config['use_diff']:
+            # Exclude the samples labeled as difficult
+            non_diff_objs = [
+                obj for obj in objs if int(obj.find('difficult').text) == 0]
+            # if len(non_diff_objs) != len(objs):
+            #     print 'Removed {} difficult objects'.format(
+            #         len(objs) - len(non_diff_objs))
+            objs = non_diff_objs
+        num_objs = len(objs)
+
+        boxes = np.zeros((num_objs, 4), dtype=np.uint16)
+        gt_classes = np.zeros((num_objs), dtype=np.int32)
+        overlaps = np.zeros((num_objs, self.num_classes), dtype=np.float32)
+        # "Seg" area for pascal is just the box area
+        seg_areas = np.zeros((num_objs), dtype=np.float32)
+
+        # Load object bounding boxes into a data frame.
+        for ix, obj in enumerate(objs):
+            bbox = obj.find('bndbox')
+            # Make pixel indexes 0-based
+            x1 = float(bbox.find('xmin').text) - 1
+            y1 = float(bbox.find('ymin').text) - 1
+            x2 = float(bbox.find('xmax').text) - 1
+            y2 = float(bbox.find('ymax').text) - 1
+            cls = self._class_to_ind[obj.find('name').text.lower().strip()]
+            boxes[ix, :] = [x1, y1, x2, y2]
+            gt_classes[ix] = cls
+            overlaps[ix, cls] = 1.0
+            seg_areas[ix] = (x2 - x1 + 1) * (y2 - y1 + 1)
+
+        overlaps = scipy.sparse.csr_matrix(overlaps)
+
+        return {'boxes' : boxes,
+                'gt_classes': gt_classes,
+                'gt_overlaps' : overlaps,
+                'flipped' : False,
+                'seg_areas' : seg_areas}
+
+    def _get_comp_id(self):
+        comp_id = (self._comp_id + '_' + self._salt if self.config['use_salt']
+            else self._comp_id)
+        return comp_id
+
+    def _get_voc_results_file_template(self):
+        # VOCdevkit/results/VOC2007/Main/<comp_id>_det_test_aeroplane.txt
+        filename = self._get_comp_id() + '_det_' + self._image_set + '_{:s}.txt'
+        path = os.path.join(
+            self._devkit_path,
+            'results',
+            'VOC' + self._year,
+            'Main',
+            filename)
+        return path
+
+    def _write_voc_results_file(self, all_boxes):
+        for cls_ind, cls in enumerate(self.classes):
+            if cls == '__background__':
+                continue
+            print 'Writing {} VOC results file'.format(cls)
+            filename = self._get_voc_results_file_template().format(cls)
+            with open(filename, 'wt') as f:
+                for im_ind, index in enumerate(self.image_index):
+                    dets = all_boxes[cls_ind][im_ind]
+                    if dets == []:
+                        continue
+                    # the VOCdevkit expects 1-based indices
+                    for k in xrange(dets.shape[0]):
+                        f.write('{:s} {:.3f} {:.1f} {:.1f} {:.1f} {:.1f}\n'.
+                                format(index, dets[k, -1],
+                                       dets[k, 0] + 1, dets[k, 1] + 1,
+                                       dets[k, 2] + 1, dets[k, 3] + 1))
+
+    def _do_python_eval(self, output_dir = 'output'):
+        annopath = os.path.join(
+            self._devkit_path,
+            'VOC' + self._year,
+            'Annotations',
+            '{:s}.xml')
+        imagesetfile = os.path.join(
+            self._devkit_path,
+            'VOC' + self._year,
+            'ImageSets',
+            'Main',
+            self._image_set + '.txt')
+        cachedir = os.path.join(self._devkit_path, 'annotations_cache')
+        aps = []
+        # The PASCAL VOC metric changed in 2010
+        use_07_metric = True if int(self._year) < 2010 else False
+        print 'VOC07 metric? ' + ('Yes' if use_07_metric else 'No')
+        if not os.path.isdir(output_dir):
+            os.mkdir(output_dir)
+        for i, cls in enumerate(self._classes):
+            if cls == '__background__':
+                continue
+            filename = self._get_voc_results_file_template().format(cls)
+            rec, prec, ap = voc_eval(
+                filename, annopath, imagesetfile, cls, cachedir, ovthresh=0.5,
+                use_07_metric=use_07_metric)
+            aps += [ap]
+            print('AP for {} = {:.4f}'.format(cls, ap))
+            with open(os.path.join(output_dir, cls + '_pr.pkl'), 'w') as f:
+                cPickle.dump({'rec': rec, 'prec': prec, 'ap': ap}, f)
+        print('Mean AP = {:.4f}'.format(np.mean(aps)))
+        print('~~~~~~~~')
+        print('Results:')
+        for ap in aps:
+            print('{:.3f}'.format(ap))
+        print('{:.3f}'.format(np.mean(aps)))
+        print('~~~~~~~~')
+        print('')
+        print('--------------------------------------------------------------')
+        print('Results computed with the **unofficial** Python eval code.')
+        print('Results should be very close to the official MATLAB eval code.')
+        print('Recompute with `./tools/reval.py --matlab ...` for your paper.')
+        print('-- Thanks, The Management')
+        print('--------------------------------------------------------------')
+
+    def _do_matlab_eval(self, output_dir='output'):
+        print '-----------------------------------------------------'
+        print 'Computing results with the official MATLAB eval code.'
+        print '-----------------------------------------------------'
+        path = os.path.join(cfg.ROOT_DIR, 'lib', 'datasets',
+                            'VOCdevkit-matlab-wrapper')
+        cmd = 'cd {} && '.format(path)
+        cmd += '{:s} -nodisplay -nodesktop '.format(cfg.MATLAB)
+        cmd += '-r "dbstop if error; '
+        cmd += 'voc_eval(\'{:s}\',\'{:s}\',\'{:s}\',\'{:s}\'); quit;"' \
+               .format(self._devkit_path, self._get_comp_id(),
+                       self._image_set, output_dir)
+        print('Running:\n{}'.format(cmd))
+        status = subprocess.call(cmd, shell=True)
+
+    def evaluate_detections(self, all_boxes, output_dir):
+        self._write_voc_results_file(all_boxes)
+        self._do_python_eval(output_dir)
+        if self.config['matlab_eval']:
+            self._do_matlab_eval(output_dir)
+        if self.config['cleanup']:
+            for cls in self._classes:
+                if cls == '__background__':
+                    continue
+                filename = self._get_voc_results_file_template().format(cls)
+                os.remove(filename)
+
+    def competition_mode(self, on):
+        if on:
+            self.config['use_salt'] = False
+            self.config['cleanup'] = False
+        else:
+            self.config['use_salt'] = True
+            self.config['cleanup'] = True
+
+if __name__ == '__main__':
+    from datasets.pascal_voc import pascal_voc
+    d = pascal_voc('trainval', '2007')
+    res = d.roidb
+    from IPython import embed; embed()
diff --git a/lib/datasets/tools/mcg_munge.py b/lib/datasets/tools/mcg_munge.py
new file mode 100644
index 0000000..1392aa3
--- /dev/null
+++ b/lib/datasets/tools/mcg_munge.py
@@ -0,0 +1,38 @@
+import os
+import sys
+
+"""Hacky tool to convert file system layout of MCG boxes downloaded from
+http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/mcg/
+so that it's consistent with those computed by Jan Hosang (see:
+http://www.mpi-inf.mpg.de/departments/computer-vision-and-multimodal-
+  computing/research/object-recognition-and-scene-understanding/how-
+  good-are-detection-proposals-really/)
+
+NB: Boxes from the MCG website are in (y1, x1, y2, x2) order.
+Boxes from Hosang et al. are in (x1, y1, x2, y2) order.
+"""
+
+def munge(src_dir):
+    # stored as: ./MCG-COCO-val2014-boxes/COCO_val2014_000000193401.mat
+    # want:      ./MCG/mat/COCO_val2014_0/COCO_val2014_000000141/COCO_val2014_000000141334.mat
+
+    files = os.listdir(src_dir)
+    for fn in files:
+        base, ext = os.path.splitext(fn)
+        # first 14 chars / first 22 chars / all chars + .mat
+        # COCO_val2014_0/COCO_val2014_000000447/COCO_val2014_000000447991.mat
+        first = base[:14]
+        second = base[:22]
+        dst_dir = os.path.join('MCG', 'mat', first, second)
+        if not os.path.exists(dst_dir):
+            os.makedirs(dst_dir)
+        src = os.path.join(src_dir, fn)
+        dst = os.path.join(dst_dir, fn)
+        print 'MV: {} -> {}'.format(src, dst)
+        os.rename(src, dst)
+
+if __name__ == '__main__':
+    # src_dir should look something like:
+    #  src_dir = 'MCG-COCO-val2014-boxes'
+    src_dir = sys.argv[1]
+    munge(src_dir)
diff --git a/lib/datasets/voc_eval.py b/lib/datasets/voc_eval.py
new file mode 100644
index 0000000..8d0a830
--- /dev/null
+++ b/lib/datasets/voc_eval.py
@@ -0,0 +1,200 @@
+# --------------------------------------------------------
+# Fast/er R-CNN
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Bharath Hariharan
+# --------------------------------------------------------
+
+import xml.etree.ElementTree as ET
+import os
+import cPickle
+import numpy as np
+
+def parse_rec(filename):
+    """ Parse a PASCAL VOC xml file """
+    tree = ET.parse(filename)
+    objects = []
+    for obj in tree.findall('object'):
+        obj_struct = {}
+        obj_struct['name'] = obj.find('name').text
+        obj_struct['pose'] = obj.find('pose').text
+        obj_struct['truncated'] = int(obj.find('truncated').text)
+        obj_struct['difficult'] = int(obj.find('difficult').text)
+        bbox = obj.find('bndbox')
+        obj_struct['bbox'] = [int(bbox.find('xmin').text),
+                              int(bbox.find('ymin').text),
+                              int(bbox.find('xmax').text),
+                              int(bbox.find('ymax').text)]
+        objects.append(obj_struct)
+
+    return objects
+
+def voc_ap(rec, prec, use_07_metric=False):
+    """ ap = voc_ap(rec, prec, [use_07_metric])
+    Compute VOC AP given precision and recall.
+    If use_07_metric is true, uses the
+    VOC 07 11 point method (default:False).
+    """
+    if use_07_metric:
+        # 11 point metric
+        ap = 0.
+        for t in np.arange(0., 1.1, 0.1):
+            if np.sum(rec >= t) == 0:
+                p = 0
+            else:
+                p = np.max(prec[rec >= t])
+            ap = ap + p / 11.
+    else:
+        # correct AP calculation
+        # first append sentinel values at the end
+        mrec = np.concatenate(([0.], rec, [1.]))
+        mpre = np.concatenate(([0.], prec, [0.]))
+
+        # compute the precision envelope
+        for i in range(mpre.size - 1, 0, -1):
+            mpre[i - 1] = np.maximum(mpre[i - 1], mpre[i])
+
+        # to calculate area under PR curve, look for points
+        # where X axis (recall) changes value
+        i = np.where(mrec[1:] != mrec[:-1])[0]
+
+        # and sum (\Delta recall) * prec
+        ap = np.sum((mrec[i + 1] - mrec[i]) * mpre[i + 1])
+    return ap
+
+def voc_eval(detpath,
+             annopath,
+             imagesetfile,
+             classname,
+             cachedir,
+             ovthresh=0.5,
+             use_07_metric=False):
+    """rec, prec, ap = voc_eval(detpath,
+                                annopath,
+                                imagesetfile,
+                                classname,
+                                [ovthresh],
+                                [use_07_metric])
+
+    Top level function that does the PASCAL VOC evaluation.
+
+    detpath: Path to detections
+        detpath.format(classname) should produce the detection results file.
+    annopath: Path to annotations
+        annopath.format(imagename) should be the xml annotations file.
+    imagesetfile: Text file containing the list of images, one image per line.
+    classname: Category name (duh)
+    cachedir: Directory for caching the annotations
+    [ovthresh]: Overlap threshold (default = 0.5)
+    [use_07_metric]: Whether to use VOC07's 11 point AP computation
+        (default False)
+    """
+    # assumes detections are in detpath.format(classname)
+    # assumes annotations are in annopath.format(imagename)
+    # assumes imagesetfile is a text file with each line an image name
+    # cachedir caches the annotations in a pickle file
+
+    # first load gt
+    if not os.path.isdir(cachedir):
+        os.mkdir(cachedir)
+    cachefile = os.path.join(cachedir, 'annots.pkl')
+    # read list of images
+    with open(imagesetfile, 'r') as f:
+        lines = f.readlines()
+    imagenames = [x.strip() for x in lines]
+
+    if not os.path.isfile(cachefile):
+        # load annots
+        recs = {}
+        for i, imagename in enumerate(imagenames):
+            recs[imagename] = parse_rec(annopath.format(imagename))
+            if i % 100 == 0:
+                print 'Reading annotation for {:d}/{:d}'.format(
+                    i + 1, len(imagenames))
+        # save
+        print 'Saving cached annotations to {:s}'.format(cachefile)
+        with open(cachefile, 'w') as f:
+            cPickle.dump(recs, f)
+    else:
+        # load
+        with open(cachefile, 'r') as f:
+            recs = cPickle.load(f)
+
+    # extract gt objects for this class
+    class_recs = {}
+    npos = 0
+    for imagename in imagenames:
+        R = [obj for obj in recs[imagename] if obj['name'] == classname]
+        bbox = np.array([x['bbox'] for x in R])
+        difficult = np.array([x['difficult'] for x in R]).astype(np.bool)
+        det = [False] * len(R)
+        npos = npos + sum(~difficult)
+        class_recs[imagename] = {'bbox': bbox,
+                                 'difficult': difficult,
+                                 'det': det}
+
+    # read dets
+    detfile = detpath.format(classname)
+    with open(detfile, 'r') as f:
+        lines = f.readlines()
+
+    splitlines = [x.strip().split(' ') for x in lines]
+    image_ids = [x[0] for x in splitlines]
+    confidence = np.array([float(x[1]) for x in splitlines])
+    BB = np.array([[float(z) for z in x[2:]] for x in splitlines])
+
+    # sort by confidence
+    sorted_ind = np.argsort(-confidence)
+    sorted_scores = np.sort(-confidence)
+    BB = BB[sorted_ind, :]
+    image_ids = [image_ids[x] for x in sorted_ind]
+
+    # go down dets and mark TPs and FPs
+    nd = len(image_ids)
+    tp = np.zeros(nd)
+    fp = np.zeros(nd)
+    for d in range(nd):
+        R = class_recs[image_ids[d]]
+        bb = BB[d, :].astype(float)
+        ovmax = -np.inf
+        BBGT = R['bbox'].astype(float)
+
+        if BBGT.size > 0:
+            # compute overlaps
+            # intersection
+            ixmin = np.maximum(BBGT[:, 0], bb[0])
+            iymin = np.maximum(BBGT[:, 1], bb[1])
+            ixmax = np.minimum(BBGT[:, 2], bb[2])
+            iymax = np.minimum(BBGT[:, 3], bb[3])
+            iw = np.maximum(ixmax - ixmin + 1., 0.)
+            ih = np.maximum(iymax - iymin + 1., 0.)
+            inters = iw * ih
+
+            # union
+            uni = ((bb[2] - bb[0] + 1.) * (bb[3] - bb[1] + 1.) +
+                   (BBGT[:, 2] - BBGT[:, 0] + 1.) *
+                   (BBGT[:, 3] - BBGT[:, 1] + 1.) - inters)
+
+            overlaps = inters / uni
+            ovmax = np.max(overlaps)
+            jmax = np.argmax(overlaps)
+
+        if ovmax > ovthresh:
+            if not R['difficult'][jmax]:
+                if not R['det'][jmax]:
+                    tp[d] = 1.
+                    R['det'][jmax] = 1
+                else:
+                    fp[d] = 1.
+        else:
+            fp[d] = 1.
+
+    # compute precision recall
+    fp = np.cumsum(fp)
+    tp = np.cumsum(tp)
+    rec = tp / float(npos)
+    # avoid divide by zero in case the first detection matches a difficult
+    # ground truth
+    prec = tp / np.maximum(tp + fp, np.finfo(np.float64).eps)
+    ap = voc_ap(rec, prec, use_07_metric)
+
+    return rec, prec, ap
diff --git a/lib/fast_rcnn/__init__.py b/lib/fast_rcnn/__init__.py
new file mode 100644
index 0000000..7ba6a65
--- /dev/null
+++ b/lib/fast_rcnn/__init__.py
@@ -0,0 +1,6 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
diff --git a/lib/fast_rcnn/bbox_transform.py b/lib/fast_rcnn/bbox_transform.py
new file mode 100644
index 0000000..c83109c
--- /dev/null
+++ b/lib/fast_rcnn/bbox_transform.py
@@ -0,0 +1,75 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+import numpy as np
+
+def bbox_transform(ex_rois, gt_rois):
+    ex_widths = ex_rois[:, 2] - ex_rois[:, 0] + 1.0
+    ex_heights = ex_rois[:, 3] - ex_rois[:, 1] + 1.0
+    ex_ctr_x = ex_rois[:, 0] + 0.5 * ex_widths
+    ex_ctr_y = ex_rois[:, 1] + 0.5 * ex_heights
+
+    gt_widths = gt_rois[:, 2] - gt_rois[:, 0] + 1.0
+    gt_heights = gt_rois[:, 3] - gt_rois[:, 1] + 1.0
+    gt_ctr_x = gt_rois[:, 0] + 0.5 * gt_widths
+    gt_ctr_y = gt_rois[:, 1] + 0.5 * gt_heights
+
+    targets_dx = (gt_ctr_x - ex_ctr_x) / ex_widths
+    targets_dy = (gt_ctr_y - ex_ctr_y) / ex_heights
+    targets_dw = np.log(gt_widths / ex_widths)
+    targets_dh = np.log(gt_heights / ex_heights)
+
+    targets = np.vstack(
+        (targets_dx, targets_dy, targets_dw, targets_dh)).transpose()
+    return targets
+
+def bbox_transform_inv(boxes, deltas):
+    if boxes.shape[0] == 0:
+        return np.zeros((0, deltas.shape[1]), dtype=deltas.dtype)
+    boxes = boxes.astype(deltas.dtype, copy=False)
+
+    widths = boxes[:, 2] - boxes[:, 0] + 1.0
+    heights = boxes[:, 3] - boxes[:, 1] + 1.0
+    ctr_x = boxes[:, 0] + 0.5 * widths
+    ctr_y = boxes[:, 1] + 0.5 * heights
+
+    dx = deltas[:, 0::4]
+    dy = deltas[:, 1::4]
+    dw = deltas[:, 2::4]
+    dh = deltas[:, 3::4]
+
+    pred_ctr_x = dx * widths[:, np.newaxis] + ctr_x[:, np.newaxis]
+    pred_ctr_y = dy * heights[:, np.newaxis] + ctr_y[:, np.newaxis]
+    pred_w = np.exp(dw) * widths[:, np.newaxis]
+    pred_h = np.exp(dh) * heights[:, np.newaxis]
+
+    pred_boxes = np.zeros(deltas.shape, dtype=deltas.dtype)
+    # x1
+    pred_boxes[:, 0::4] = pred_ctr_x - 0.5 * pred_w
+    # y1
+    pred_boxes[:, 1::4] = pred_ctr_y - 0.5 * pred_h
+    # x2
+    pred_boxes[:, 2::4] = pred_ctr_x + 0.5 * pred_w
+    # y2
+    pred_boxes[:, 3::4] = pred_ctr_y + 0.5 * pred_h
+
+    return pred_boxes
+
+def clip_boxes(boxes, im_shape):
+    """
+    Clip boxes to image boundaries.
+    """
+
+    # x1 >= 0
+    boxes[:, 0::4] = np.maximum(np.minimum(boxes[:, 0::4], im_shape[1] - 1), 0)
+    # y1 >= 0
+    boxes[:, 1::4] = np.maximum(np.minimum(boxes[:, 1::4], im_shape[0] - 1), 0)
+    # x2 < im_shape[1]
+    boxes[:, 2::4] = np.maximum(np.minimum(boxes[:, 2::4], im_shape[1] - 1), 0)
+    # y2 < im_shape[0]
+    boxes[:, 3::4] = np.maximum(np.minimum(boxes[:, 3::4], im_shape[0] - 1), 0)
+    return boxes
diff --git a/lib/fast_rcnn/config.py b/lib/fast_rcnn/config.py
new file mode 100644
index 0000000..1751856
--- /dev/null
+++ b/lib/fast_rcnn/config.py
@@ -0,0 +1,290 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Fast R-CNN config system.
+
+This file specifies default config options for Fast R-CNN. You should not
+change values in this file. Instead, you should write a config file (in yaml)
+and use cfg_from_file(yaml_file) to load it and override the default options.
+
+Most tools in $ROOT/tools take a --cfg option to specify an override file.
+    - See tools/{train,test}_net.py for example code that uses cfg_from_file()
+    - See experiments/cfgs/*.yml for example YAML config override files
+"""
+
+import os
+import os.path as osp
+import numpy as np
+# `pip install easydict` if you don't have it
+from easydict import EasyDict as edict
+
+__C = edict()
+# Consumers can get config by:
+#   from fast_rcnn_config import cfg
+cfg = __C
+
+#
+# Training options
+#
+
+__C.TRAIN = edict()
+
+# Scales to use during training (can list multiple scales)
+# Each scale is the pixel size of an image's shortest side
+__C.TRAIN.SCALES = (600,)
+
+# Max pixel size of the longest side of a scaled input image
+__C.TRAIN.MAX_SIZE = 1000
+
+# Images to use per minibatch
+__C.TRAIN.IMS_PER_BATCH = 2
+
+# Minibatch size (number of regions of interest [ROIs])
+__C.TRAIN.BATCH_SIZE = 128
+
+# Fraction of minibatch that is labeled foreground (i.e. class > 0)
+__C.TRAIN.FG_FRACTION = 0.25
+
+# Overlap threshold for a ROI to be considered foreground (if >= FG_THRESH)
+__C.TRAIN.FG_THRESH = 0.5
+
+# Overlap threshold for a ROI to be considered background (class = 0 if
+# overlap in [LO, HI))
+__C.TRAIN.BG_THRESH_HI = 0.5
+__C.TRAIN.BG_THRESH_LO = 0.1
+
+# Use horizontally-flipped images during training?
+__C.TRAIN.USE_FLIPPED = True
+
+# Train bounding-box regressors
+__C.TRAIN.BBOX_REG = True
+
+# Overlap required between a ROI and ground-truth box in order for that ROI to
+# be used as a bounding-box regression training example
+__C.TRAIN.BBOX_THRESH = 0.5
+
+# Iterations between snapshots
+__C.TRAIN.SNAPSHOT_ITERS = 10000
+
+# solver.prototxt specifies the snapshot path prefix, this adds an optional
+# infix to yield the path: <prefix>[_<infix>]_iters_XYZ.caffemodel
+__C.TRAIN.SNAPSHOT_INFIX = ''
+
+# Use a prefetch thread in roi_data_layer.layer
+# So far I haven't found this useful; likely more engineering work is required
+__C.TRAIN.USE_PREFETCH = False
+
+# Normalize the targets (subtract empirical mean, divide by empirical stddev)
+__C.TRAIN.BBOX_NORMALIZE_TARGETS = True
+# Deprecated (inside weights)
+__C.TRAIN.BBOX_INSIDE_WEIGHTS = (1.0, 1.0, 1.0, 1.0)
+# Normalize the targets using "precomputed" (or made up) means and stdevs
+# (BBOX_NORMALIZE_TARGETS must also be True)
+__C.TRAIN.BBOX_NORMALIZE_TARGETS_PRECOMPUTED = False
+__C.TRAIN.BBOX_NORMALIZE_MEANS = (0.0, 0.0, 0.0, 0.0)
+__C.TRAIN.BBOX_NORMALIZE_STDS = (0.1, 0.1, 0.2, 0.2)
+
+# Train using these proposals
+__C.TRAIN.PROPOSAL_METHOD = 'selective_search'
+
+# Make minibatches from images that have similar aspect ratios (i.e. both
+# tall and thin or both short and wide) in order to avoid wasting computation
+# on zero-padding.
+__C.TRAIN.ASPECT_GROUPING = True
+
+# Use RPN to detect objects
+__C.TRAIN.HAS_RPN = False
+# IOU >= thresh: positive example
+__C.TRAIN.RPN_POSITIVE_OVERLAP = 0.7
+# IOU < thresh: negative example
+__C.TRAIN.RPN_NEGATIVE_OVERLAP = 0.3
+# If an anchor statisfied by positive and negative conditions set to negative
+__C.TRAIN.RPN_CLOBBER_POSITIVES = False
+# Max number of foreground examples
+__C.TRAIN.RPN_FG_FRACTION = 0.5
+# Total number of examples
+__C.TRAIN.RPN_BATCHSIZE = 256
+# NMS threshold used on RPN proposals
+__C.TRAIN.RPN_NMS_THRESH = 0.7
+# Number of top scoring boxes to keep before apply NMS to RPN proposals
+__C.TRAIN.RPN_PRE_NMS_TOP_N = 12000
+# Number of top scoring boxes to keep after applying NMS to RPN proposals
+__C.TRAIN.RPN_POST_NMS_TOP_N = 2000
+# Proposal height and width both need to be greater than RPN_MIN_SIZE (at orig image scale)
+__C.TRAIN.RPN_MIN_SIZE = 16
+# Deprecated (outside weights)
+__C.TRAIN.RPN_BBOX_INSIDE_WEIGHTS = (1.0, 1.0, 1.0, 1.0)
+# Give the positive RPN examples weight of p * 1 / {num positives}
+# and give negatives a weight of (1 - p)
+# Set to -1.0 to use uniform example weighting
+__C.TRAIN.RPN_POSITIVE_WEIGHT = -1.0
+
+# whether use class aware box or not
+__C.TRAIN.AGONISTIC = False
+
+#
+# Testing options
+#
+
+__C.TEST = edict()
+
+# Scales to use during testing (can list multiple scales)
+# Each scale is the pixel size of an image's shortest side
+__C.TEST.SCALES = (600,)
+
+# Max pixel size of the longest side of a scaled input image
+__C.TEST.MAX_SIZE = 1000
+
+# Overlap threshold used for non-maximum suppression (suppress boxes with
+# IoU >= this threshold)
+__C.TEST.NMS = 0.3
+
+# Experimental: treat the (K+1) units in the cls_score layer as linear
+# predictors (trained, eg, with one-vs-rest SVMs).
+__C.TEST.SVM = False
+
+# Test using bounding-box regressors
+__C.TEST.BBOX_REG = True
+
+# Propose boxes
+__C.TEST.HAS_RPN = False
+
+# Test using these proposals
+__C.TEST.PROPOSAL_METHOD = 'selective_search'
+
+## NMS threshold used on RPN proposals
+__C.TEST.RPN_NMS_THRESH = 0.7
+## Number of top scoring boxes to keep before apply NMS to RPN proposals
+__C.TEST.RPN_PRE_NMS_TOP_N = 6000
+## Number of top scoring boxes to keep after applying NMS to RPN proposals
+__C.TEST.RPN_POST_NMS_TOP_N = 300
+# Proposal height and width both need to be greater than RPN_MIN_SIZE (at orig image scale)
+__C.TEST.RPN_MIN_SIZE = 16
+
+# whether use class aware box or not
+__C.TEST.AGONISTIC = False
+
+
+#
+# MISC
+#
+
+# The mapping from image coordinates to feature map coordinates might cause
+# some boxes that are distinct in image space to become identical in feature
+# coordinates. If DEDUP_BOXES > 0, then DEDUP_BOXES is used as the scale factor
+# for identifying duplicate boxes.
+# 1/16 is correct for {Alex,Caffe}Net, VGG_CNN_M_1024, and VGG16
+__C.DEDUP_BOXES = 1./16.
+
+# Pixel mean values (BGR order) as a (1, 1, 3) array
+# We use the same pixel mean for all networks even though it's not exactly what
+# they were trained with
+__C.PIXEL_MEANS = np.array([[[102.9801, 115.9465, 122.7717]]])
+
+# For reproducibility
+__C.RNG_SEED = 3
+
+# A small number that's used many times
+__C.EPS = 1e-14
+
+# Root directory of project
+__C.ROOT_DIR = osp.abspath(osp.join(osp.dirname(__file__), '..', '..'))
+
+# Data directory
+__C.DATA_DIR = osp.abspath(osp.join(__C.ROOT_DIR, 'data'))
+
+# Model directory
+__C.MODELS_DIR = osp.abspath(osp.join(__C.ROOT_DIR, 'models', 'pascal_voc'))
+
+# Name (or path to) the matlab executable
+__C.MATLAB = 'matlab'
+
+# Place outputs under an experiments directory
+__C.EXP_DIR = 'default'
+
+# Use GPU implementation of non-maximum suppression
+__C.USE_GPU_NMS = True
+
+# Default GPU device id
+__C.GPU_ID = 0
+
+
+def get_output_dir(imdb, net=None):
+    """Return the directory where experimental artifacts are placed.
+    If the directory does not exist, it is created.
+
+    A canonical path is built using the name from an imdb and a network
+    (if not None).
+    """
+    outdir = osp.abspath(osp.join(__C.ROOT_DIR, 'output', __C.EXP_DIR, imdb.name))
+    if net is not None:
+        outdir = osp.join(outdir, net.name)
+    if not os.path.exists(outdir):
+        os.makedirs(outdir)
+    return outdir
+
+def _merge_a_into_b(a, b):
+    """Merge config dictionary a into config dictionary b, clobbering the
+    options in b whenever they are also specified in a.
+    """
+    if type(a) is not edict:
+        return
+
+    for k, v in a.iteritems():
+        # a must specify keys that are in b
+        if not b.has_key(k):
+            raise KeyError('{} is not a valid config key'.format(k))
+
+        # the types must match, too
+        old_type = type(b[k])
+        if old_type is not type(v):
+            if isinstance(b[k], np.ndarray):
+                v = np.array(v, dtype=b[k].dtype)
+            else:
+                raise ValueError(('Type mismatch ({} vs. {}) '
+                                'for config key: {}').format(type(b[k]),
+                                                            type(v), k))
+
+        # recursively merge dicts
+        if type(v) is edict:
+            try:
+                _merge_a_into_b(a[k], b[k])
+            except:
+                print('Error under config key: {}'.format(k))
+                raise
+        else:
+            b[k] = v
+
+def cfg_from_file(filename):
+    """Load a config file and merge it into the default options."""
+    import yaml
+    with open(filename, 'r') as f:
+        yaml_cfg = edict(yaml.load(f))
+
+    _merge_a_into_b(yaml_cfg, __C)
+
+def cfg_from_list(cfg_list):
+    """Set config keys via list (e.g., from command line)."""
+    from ast import literal_eval
+    assert len(cfg_list) % 2 == 0
+    for k, v in zip(cfg_list[0::2], cfg_list[1::2]):
+        key_list = k.split('.')
+        d = __C
+        for subkey in key_list[:-1]:
+            assert d.has_key(subkey)
+            d = d[subkey]
+        subkey = key_list[-1]
+        assert d.has_key(subkey)
+        try:
+            value = literal_eval(v)
+        except:
+            # handle the case when v is a string literal
+            value = v
+        assert type(value) == type(d[subkey]), \
+            'type {} does not match original type {}'.format(
+            type(value), type(d[subkey]))
+        d[subkey] = value
diff --git a/lib/fast_rcnn/nms_wrapper.py b/lib/fast_rcnn/nms_wrapper.py
new file mode 100644
index 0000000..d1a11db
--- /dev/null
+++ b/lib/fast_rcnn/nms_wrapper.py
@@ -0,0 +1,20 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+from fast_rcnn.config import cfg
+from nms.gpu_nms import gpu_nms
+from nms.cpu_nms import cpu_nms
+
+def nms(dets, thresh, force_cpu=False):
+    """Dispatch to either CPU or GPU NMS implementations."""
+
+    if dets.shape[0] == 0:
+        return []
+    if cfg.USE_GPU_NMS and not force_cpu:
+        return gpu_nms(dets, thresh, device_id=cfg.GPU_ID)
+    else:
+        return cpu_nms(dets, thresh)
diff --git a/lib/fast_rcnn/test.py b/lib/fast_rcnn/test.py
new file mode 100644
index 0000000..78c24ae
--- /dev/null
+++ b/lib/fast_rcnn/test.py
@@ -0,0 +1,298 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Test a Fast R-CNN network on an imdb (image database)."""
+
+from fast_rcnn.config import cfg, get_output_dir
+from fast_rcnn.bbox_transform import clip_boxes, bbox_transform_inv
+import argparse
+from utils.timer import Timer
+import numpy as np
+import cv2
+import caffe
+from fast_rcnn.nms_wrapper import nms
+import cPickle
+from utils.blob import im_list_to_blob
+import os
+
+def _get_image_blob(im):
+    """Converts an image into a network input.
+
+    Arguments:
+        im (ndarray): a color image in BGR order
+
+    Returns:
+        blob (ndarray): a data blob holding an image pyramid
+        im_scale_factors (list): list of image scales (relative to im) used
+            in the image pyramid
+    """
+    im_orig = im.astype(np.float32, copy=True)
+    im_orig -= cfg.PIXEL_MEANS
+
+    im_shape = im_orig.shape
+    im_size_min = np.min(im_shape[0:2])
+    im_size_max = np.max(im_shape[0:2])
+
+    processed_ims = []
+    im_scale_factors = []
+
+    for target_size in cfg.TEST.SCALES:
+        im_scale = float(target_size) / float(im_size_min)
+        # Prevent the biggest axis from being more than MAX_SIZE
+        if np.round(im_scale * im_size_max) > cfg.TEST.MAX_SIZE:
+            im_scale = float(cfg.TEST.MAX_SIZE) / float(im_size_max)
+        im = cv2.resize(im_orig, None, None, fx=im_scale, fy=im_scale,
+                        interpolation=cv2.INTER_LINEAR)
+        im_scale_factors.append(im_scale)
+        processed_ims.append(im)
+
+    # Create a blob to hold the input images
+    blob = im_list_to_blob(processed_ims)
+
+    return blob, np.array(im_scale_factors)
+
+def _get_rois_blob(im_rois, im_scale_factors):
+    """Converts RoIs into network inputs.
+
+    Arguments:
+        im_rois (ndarray): R x 4 matrix of RoIs in original image coordinates
+        im_scale_factors (list): scale factors as returned by _get_image_blob
+
+    Returns:
+        blob (ndarray): R x 5 matrix of RoIs in the image pyramid
+    """
+    rois, levels = _project_im_rois(im_rois, im_scale_factors)
+    rois_blob = np.hstack((levels, rois))
+    return rois_blob.astype(np.float32, copy=False)
+
+def _project_im_rois(im_rois, scales):
+    """Project image RoIs into the image pyramid built by _get_image_blob.
+
+    Arguments:
+        im_rois (ndarray): R x 4 matrix of RoIs in original image coordinates
+        scales (list): scale factors as returned by _get_image_blob
+
+    Returns:
+        rois (ndarray): R x 4 matrix of projected RoI coordinates
+        levels (list): image pyramid levels used by each projected RoI
+    """
+    im_rois = im_rois.astype(np.float, copy=False)
+
+    if len(scales) > 1:
+        widths = im_rois[:, 2] - im_rois[:, 0] + 1
+        heights = im_rois[:, 3] - im_rois[:, 1] + 1
+
+        areas = widths * heights
+        scaled_areas = areas[:, np.newaxis] * (scales[np.newaxis, :] ** 2)
+        diff_areas = np.abs(scaled_areas - 224 * 224)
+        levels = diff_areas.argmin(axis=1)[:, np.newaxis]
+    else:
+        levels = np.zeros((im_rois.shape[0], 1), dtype=np.int)
+
+    rois = im_rois * scales[levels]
+
+    return rois, levels
+
+def _get_blobs(im, rois):
+    """Convert an image and RoIs within that image into network inputs."""
+    blobs = {'data' : None, 'rois' : None}
+    blobs['data'], im_scale_factors = _get_image_blob(im)
+    if not cfg.TEST.HAS_RPN:
+        blobs['rois'] = _get_rois_blob(rois, im_scale_factors)
+    return blobs, im_scale_factors
+
+def im_detect(net, im, boxes=None):
+    """Detect object classes in an image given object proposals.
+
+    Arguments:
+        net (caffe.Net): Fast R-CNN network to use
+        im (ndarray): color image to test (in BGR order)
+        boxes (ndarray): R x 4 array of object proposals or None (for RPN)
+
+    Returns:
+        scores (ndarray): R x K array of object class scores (K includes
+            background as object category 0)
+        boxes (ndarray): R x (4*K) array of predicted bounding boxes
+    """
+    blobs, im_scales = _get_blobs(im, boxes)
+
+    # When mapping from image ROIs to feature map ROIs, there's some aliasing
+    # (some distinct image ROIs get mapped to the same feature ROI).
+    # Here, we identify duplicate feature ROIs, so we only compute features
+    # on the unique subset.
+    if cfg.DEDUP_BOXES > 0 and not cfg.TEST.HAS_RPN:
+        v = np.array([1, 1e3, 1e6, 1e9, 1e12])
+        hashes = np.round(blobs['rois'] * cfg.DEDUP_BOXES).dot(v)
+        _, index, inv_index = np.unique(hashes, return_index=True,
+                                        return_inverse=True)
+        blobs['rois'] = blobs['rois'][index, :]
+        boxes = boxes[index, :]
+
+    if cfg.TEST.HAS_RPN:
+        im_blob = blobs['data']
+        blobs['im_info'] = np.array(
+            [[im_blob.shape[2], im_blob.shape[3], im_scales[0]]],
+            dtype=np.float32)
+
+    # reshape network inputs
+    net.blobs['data'].reshape(*(blobs['data'].shape))
+    if cfg.TEST.HAS_RPN:
+        net.blobs['im_info'].reshape(*(blobs['im_info'].shape))
+    else:
+        net.blobs['rois'].reshape(*(blobs['rois'].shape))
+
+    # do forward
+    forward_kwargs = {'data': blobs['data'].astype(np.float32, copy=False)}
+    if cfg.TEST.HAS_RPN:
+        forward_kwargs['im_info'] = blobs['im_info'].astype(np.float32, copy=False)
+    else:
+        forward_kwargs['rois'] = blobs['rois'].astype(np.float32, copy=False)
+    blobs_out = net.forward(**forward_kwargs)
+
+    if cfg.TEST.HAS_RPN:
+        assert len(im_scales) == 1, "Only single-image batch implemented"
+        rois = net.blobs['rois'].data.copy()
+        # unscale back to raw image space
+        boxes = rois[:, 1:5] / im_scales[0]
+
+    if cfg.TEST.SVM:
+        # use the raw scores before softmax under the assumption they
+        # were trained as linear SVMs
+        scores = net.blobs['cls_score'].data
+    else:
+        # use softmax estimated probabilities
+        scores = blobs_out['cls_prob']
+
+    if cfg.TEST.BBOX_REG:
+        # Apply bounding-box regression deltas
+        box_deltas = blobs_out['bbox_pred']
+        pred_boxes = bbox_transform_inv(boxes, box_deltas)
+        pred_boxes = clip_boxes(pred_boxes, im.shape)
+    else:
+        # Simply repeat the boxes, once for each class
+        pred_boxes = np.tile(boxes, (1, scores.shape[1]))
+
+    if cfg.DEDUP_BOXES > 0 and not cfg.TEST.HAS_RPN:
+        # Map scores and predictions back to the original set of boxes
+        scores = scores[inv_index, :]
+        pred_boxes = pred_boxes[inv_index, :]
+
+    return scores, pred_boxes
+
+def vis_detections(im, class_name, dets, thresh=0.3):
+    """Visual debugging of detections."""
+    import matplotlib.pyplot as plt
+    im = im[:, :, (2, 1, 0)]
+    for i in xrange(np.minimum(10, dets.shape[0])):
+        bbox = dets[i, :4]
+        score = dets[i, -1]
+        if score > thresh:
+            plt.cla()
+            plt.imshow(im)
+            plt.gca().add_patch(
+                plt.Rectangle((bbox[0], bbox[1]),
+                              bbox[2] - bbox[0],
+                              bbox[3] - bbox[1], fill=False,
+                              edgecolor='g', linewidth=3)
+                )
+            plt.title('{}  {:.3f}'.format(class_name, score))
+            plt.show()
+
+def apply_nms(all_boxes, thresh):
+    """Apply non-maximum suppression to all predicted boxes output by the
+    test_net method.
+    """
+    num_classes = len(all_boxes)
+    num_images = len(all_boxes[0])
+    nms_boxes = [[[] for _ in xrange(num_images)]
+                 for _ in xrange(num_classes)]
+    for cls_ind in xrange(num_classes):
+        for im_ind in xrange(num_images):
+            dets = all_boxes[cls_ind][im_ind]
+            if dets == []:
+                continue
+            # CPU NMS is much faster than GPU NMS when the number of boxes
+            # is relative small (e.g., < 10k)
+            # TODO(rbg): autotune NMS dispatch
+            keep = nms(dets, thresh, force_cpu=True)
+            if len(keep) == 0:
+                continue
+            nms_boxes[cls_ind][im_ind] = dets[keep, :].copy()
+    return nms_boxes
+
+def test_net(net, imdb, max_per_image=100, thresh=0.05, vis=False):
+    """Test a Fast R-CNN network on an image database."""
+    num_images = len(imdb.image_index)
+    # all detections are collected into:
+    #    all_boxes[cls][image] = N x 5 array of detections in
+    #    (x1, y1, x2, y2, score)
+    all_boxes = [[[] for _ in xrange(num_images)]
+                 for _ in xrange(imdb.num_classes)]
+
+    output_dir = get_output_dir(imdb, net)
+
+    # timers
+    _t = {'im_detect' : Timer(), 'misc' : Timer()}
+
+    if not cfg.TEST.HAS_RPN:
+        roidb = imdb.roidb
+
+    for i in xrange(num_images):
+        # filter out any ground truth boxes
+        if cfg.TEST.HAS_RPN:
+            box_proposals = None
+        else:
+            # The roidb may contain ground-truth rois (for example, if the roidb
+            # comes from the training or val split). We only want to evaluate
+            # detection on the *non*-ground-truth rois. We select those the rois
+            # that have the gt_classes field set to 0, which means there's no
+            # ground truth.
+            box_proposals = roidb[i]['boxes'][roidb[i]['gt_classes'] == 0]
+
+        im = cv2.imread(imdb.image_path_at(i))
+        _t['im_detect'].tic()
+        scores, boxes = im_detect(net, im, box_proposals)
+        _t['im_detect'].toc()
+
+        _t['misc'].tic()
+        # skip j = 0, because it's the background class
+        for j in xrange(1, imdb.num_classes):
+            inds = np.where(scores[:, j] > thresh)[0]
+            cls_scores = scores[inds, j]
+            if cfg.TEST.AGONISTIC:
+                cls_boxes = boxes[inds, 4:8]
+            else:
+                cls_boxes = boxes[inds, j*4:(j+1)*4]
+            cls_dets = np.hstack((cls_boxes, cls_scores[:, np.newaxis])) \
+                .astype(np.float32, copy=False)
+            keep = nms(cls_dets, cfg.TEST.NMS)
+            cls_dets = cls_dets[keep, :]
+            if vis:
+                vis_detections(im, imdb.classes[j], cls_dets)
+            all_boxes[j][i] = cls_dets
+
+        # Limit to max_per_image detections *over all classes*
+        if max_per_image > 0:
+            image_scores = np.hstack([all_boxes[j][i][:, -1]
+                                      for j in xrange(1, imdb.num_classes)])
+            if len(image_scores) > max_per_image:
+                image_thresh = np.sort(image_scores)[-max_per_image]
+                for j in xrange(1, imdb.num_classes):
+                    keep = np.where(all_boxes[j][i][:, -1] >= image_thresh)[0]
+                    all_boxes[j][i] = all_boxes[j][i][keep, :]
+        _t['misc'].toc()
+
+        print 'im_detect: {:d}/{:d} {:.3f}s {:.3f}s' \
+              .format(i + 1, num_images, _t['im_detect'].average_time,
+                      _t['misc'].average_time)
+
+    det_file = os.path.join(output_dir, 'detections.pkl')
+    with open(det_file, 'wb') as f:
+        cPickle.dump(all_boxes, f, cPickle.HIGHEST_PROTOCOL)
+
+    print 'Evaluating detections'
+    imdb.evaluate_detections(all_boxes, output_dir)
diff --git a/lib/fast_rcnn/train.py b/lib/fast_rcnn/train.py
new file mode 100644
index 0000000..6ca5437
--- /dev/null
+++ b/lib/fast_rcnn/train.py
@@ -0,0 +1,186 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Train a Fast R-CNN network."""
+
+import caffe
+from fast_rcnn.config import cfg
+import roi_data_layer.roidb as rdl_roidb
+from utils.timer import Timer
+import numpy as np
+import os
+
+from caffe.proto import caffe_pb2
+import google.protobuf as pb2
+
+class SolverWrapper(object):
+    """A simple wrapper around Caffe's solver.
+    This wrapper gives us control over he snapshotting process, which we
+    use to unnormalize the learned bounding-box regression weights.
+    """
+
+    def __init__(self, solver_prototxt, roidb, output_dir,
+                 pretrained_model=None):
+        """Initialize the SolverWrapper."""
+        self.output_dir = output_dir
+
+        if (cfg.TRAIN.HAS_RPN and cfg.TRAIN.BBOX_REG and
+            cfg.TRAIN.BBOX_NORMALIZE_TARGETS):
+            # RPN can only use precomputed normalization because there are no
+            # fixed statistics to compute a priori
+            assert cfg.TRAIN.BBOX_NORMALIZE_TARGETS_PRECOMPUTED
+
+        if cfg.TRAIN.BBOX_REG:
+            print 'Computing bounding-box regression targets...'
+            self.bbox_means, self.bbox_stds = \
+                    rdl_roidb.add_bbox_regression_targets(roidb)
+            print 'done'
+
+        self.solver = caffe.SGDSolver(solver_prototxt)
+        if pretrained_model is not None:
+            print ('Loading pretrained model '
+                   'weights from {:s}').format(pretrained_model)
+            self.solver.net.copy_from(pretrained_model)
+
+        self.solver_param = caffe_pb2.SolverParameter()
+        with open(solver_prototxt, 'rt') as f:
+            pb2.text_format.Merge(f.read(), self.solver_param)
+
+        self.solver.net.layers[0].set_roidb(roidb)
+
+    def snapshot(self):
+        """Take a snapshot of the network after unnormalizing the learned
+        bounding-box regression weights. This enables easy use at test-time.
+        """
+        net = self.solver.net
+
+        scale_bbox_params_faster_rcnn = (cfg.TRAIN.BBOX_REG and
+                             cfg.TRAIN.BBOX_NORMALIZE_TARGETS and
+                             net.params.has_key('bbox_pred'))
+
+        scale_bbox_params_rfcn = (cfg.TRAIN.BBOX_REG and
+                             cfg.TRAIN.BBOX_NORMALIZE_TARGETS and
+                             net.params.has_key('rfcn_bbox'))
+
+        if scale_bbox_params_faster_rcnn:
+            # save original values
+            orig_0 = net.params['bbox_pred'][0].data.copy()
+            orig_1 = net.params['bbox_pred'][1].data.copy()
+
+            # scale and shift with bbox reg unnormalization; then save snapshot
+            net.params['bbox_pred'][0].data[...] = \
+                    (net.params['bbox_pred'][0].data *
+                     self.bbox_stds[:, np.newaxis])
+            net.params['bbox_pred'][1].data[...] = \
+                    (net.params['bbox_pred'][1].data *
+                     self.bbox_stds + self.bbox_means)
+
+
+        if scale_bbox_params_rfcn:
+            # save original values
+            orig_0 = net.params['rfcn_bbox'][0].data.copy()
+            orig_1 = net.params['rfcn_bbox'][1].data.copy()
+            repeat = orig_1.shape[0] / self.bbox_means.shape[0]
+                        
+
+            # scale and shift with bbox reg unnormalization; then save snapshot
+            net.params['rfcn_bbox'][0].data[...] = \
+                    (net.params['rfcn_bbox'][0].data *
+                     np.repeat(self.bbox_stds, repeat).reshape((orig_1.shape[0], 1, 1, 1)))
+            net.params['rfcn_bbox'][1].data[...] = \
+                    (net.params['rfcn_bbox'][1].data *
+                     np.repeat(self.bbox_stds, repeat) + np.repeat(self.bbox_means, repeat))
+
+        infix = ('_' + cfg.TRAIN.SNAPSHOT_INFIX
+                 if cfg.TRAIN.SNAPSHOT_INFIX != '' else '')
+        filename = (self.solver_param.snapshot_prefix + infix +
+                    '_iter_{:d}'.format(self.solver.iter) + '.caffemodel')
+        filename = os.path.join(self.output_dir, filename)
+        net.save(str(filename))
+        print 'Wrote snapshot to: {:s}'.format(filename)
+
+        if scale_bbox_params_faster_rcnn:
+            # restore net to original state
+            net.params['bbox_pred'][0].data[...] = orig_0
+            net.params['bbox_pred'][1].data[...] = orig_1
+        if scale_bbox_params_rfcn:
+            # restore net to original state
+            net.params['rfcn_bbox'][0].data[...] = orig_0
+            net.params['rfcn_bbox'][1].data[...] = orig_1
+
+        return filename
+
+    def train_model(self, max_iters):
+        """Network training loop."""
+        last_snapshot_iter = -1
+        timer = Timer()
+        model_paths = []
+        while self.solver.iter < max_iters:
+            # Make one SGD update
+            timer.tic()
+            self.solver.step(1)
+            timer.toc()
+            if self.solver.iter % (10 * self.solver_param.display) == 0:
+                print 'speed: {:.3f}s / iter'.format(timer.average_time)
+
+            if self.solver.iter % cfg.TRAIN.SNAPSHOT_ITERS == 0:
+                last_snapshot_iter = self.solver.iter
+                model_paths.append(self.snapshot())
+
+        if last_snapshot_iter != self.solver.iter:
+            model_paths.append(self.snapshot())
+        return model_paths
+
+def get_training_roidb(imdb):
+    """Returns a roidb (Region of Interest database) for use in training."""
+    if cfg.TRAIN.USE_FLIPPED:
+        print 'Appending horizontally-flipped training examples...'
+        imdb.append_flipped_images()
+        print 'done'
+
+    print 'Preparing training data...'
+    rdl_roidb.prepare_roidb(imdb)
+    print 'done'
+
+    return imdb.roidb
+
+def filter_roidb(roidb):
+    """Remove roidb entries that have no usable RoIs."""
+
+    def is_valid(entry):
+        # Valid images have:
+        #   (1) At least one foreground RoI OR
+        #   (2) At least one background RoI
+        overlaps = entry['max_overlaps']
+        # find boxes with sufficient overlap
+        fg_inds = np.where(overlaps >= cfg.TRAIN.FG_THRESH)[0]
+        # Select background RoIs as those within [BG_THRESH_LO, BG_THRESH_HI)
+        bg_inds = np.where((overlaps < cfg.TRAIN.BG_THRESH_HI) &
+                           (overlaps >= cfg.TRAIN.BG_THRESH_LO))[0]
+        # image is only valid if such boxes exist
+        valid = len(fg_inds) > 0 or len(bg_inds) > 0
+        return valid
+
+    num = len(roidb)
+    filtered_roidb = [entry for entry in roidb if is_valid(entry)]
+    num_after = len(filtered_roidb)
+    print 'Filtered {} roidb entries: {} -> {}'.format(num - num_after,
+                                                       num, num_after)
+    return filtered_roidb
+
+def train_net(solver_prototxt, roidb, output_dir,
+              pretrained_model=None, max_iters=40000):
+    """Train a Fast R-CNN network."""
+
+    roidb = filter_roidb(roidb)
+    sw = SolverWrapper(solver_prototxt, roidb, output_dir,
+                       pretrained_model=pretrained_model)
+
+    print 'Solving...'
+    model_paths = sw.train_model(max_iters)
+    print 'done solving'
+    return model_paths
diff --git a/lib/nms/.gitignore b/lib/nms/.gitignore
new file mode 100644
index 0000000..15a165d
--- /dev/null
+++ b/lib/nms/.gitignore
@@ -0,0 +1,3 @@
+*.c
+*.cpp
+*.so
diff --git a/lib/nms/__init__.py b/lib/nms/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/lib/nms/cpu_nms.pyx b/lib/nms/cpu_nms.pyx
new file mode 100644
index 0000000..1d0bef3
--- /dev/null
+++ b/lib/nms/cpu_nms.pyx
@@ -0,0 +1,68 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+import numpy as np
+cimport numpy as np
+
+cdef inline np.float32_t max(np.float32_t a, np.float32_t b):
+    return a if a >= b else b
+
+cdef inline np.float32_t min(np.float32_t a, np.float32_t b):
+    return a if a <= b else b
+
+def cpu_nms(np.ndarray[np.float32_t, ndim=2] dets, np.float thresh):
+    cdef np.ndarray[np.float32_t, ndim=1] x1 = dets[:, 0]
+    cdef np.ndarray[np.float32_t, ndim=1] y1 = dets[:, 1]
+    cdef np.ndarray[np.float32_t, ndim=1] x2 = dets[:, 2]
+    cdef np.ndarray[np.float32_t, ndim=1] y2 = dets[:, 3]
+    cdef np.ndarray[np.float32_t, ndim=1] scores = dets[:, 4]
+
+    cdef np.ndarray[np.float32_t, ndim=1] areas = (x2 - x1 + 1) * (y2 - y1 + 1)
+    cdef np.ndarray[np.int_t, ndim=1] order = scores.argsort()[::-1]
+
+    cdef int ndets = dets.shape[0]
+    cdef np.ndarray[np.int_t, ndim=1] suppressed = \
+            np.zeros((ndets), dtype=np.int)
+
+    # nominal indices
+    cdef int _i, _j
+    # sorted indices
+    cdef int i, j
+    # temp variables for box i's (the box currently under consideration)
+    cdef np.float32_t ix1, iy1, ix2, iy2, iarea
+    # variables for computing overlap with box j (lower scoring box)
+    cdef np.float32_t xx1, yy1, xx2, yy2
+    cdef np.float32_t w, h
+    cdef np.float32_t inter, ovr
+
+    keep = []
+    for _i in range(ndets):
+        i = order[_i]
+        if suppressed[i] == 1:
+            continue
+        keep.append(i)
+        ix1 = x1[i]
+        iy1 = y1[i]
+        ix2 = x2[i]
+        iy2 = y2[i]
+        iarea = areas[i]
+        for _j in range(_i + 1, ndets):
+            j = order[_j]
+            if suppressed[j] == 1:
+                continue
+            xx1 = max(ix1, x1[j])
+            yy1 = max(iy1, y1[j])
+            xx2 = min(ix2, x2[j])
+            yy2 = min(iy2, y2[j])
+            w = max(0.0, xx2 - xx1 + 1)
+            h = max(0.0, yy2 - yy1 + 1)
+            inter = w * h
+            ovr = inter / (iarea + areas[j] - inter)
+            if ovr >= thresh:
+                suppressed[j] = 1
+
+    return keep
diff --git a/lib/nms/gpu_nms.hpp b/lib/nms/gpu_nms.hpp
new file mode 100644
index 0000000..68b6d42
--- /dev/null
+++ b/lib/nms/gpu_nms.hpp
@@ -0,0 +1,2 @@
+void _nms(int* keep_out, int* num_out, const float* boxes_host, int boxes_num,
+          int boxes_dim, float nms_overlap_thresh, int device_id);
diff --git a/lib/nms/gpu_nms.pyx b/lib/nms/gpu_nms.pyx
new file mode 100644
index 0000000..59d84af
--- /dev/null
+++ b/lib/nms/gpu_nms.pyx
@@ -0,0 +1,31 @@
+# --------------------------------------------------------
+# Faster R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+import numpy as np
+cimport numpy as np
+
+assert sizeof(int) == sizeof(np.int32_t)
+
+cdef extern from "gpu_nms.hpp":
+    void _nms(np.int32_t*, int*, np.float32_t*, int, int, float, int)
+
+def gpu_nms(np.ndarray[np.float32_t, ndim=2] dets, np.float thresh,
+            np.int32_t device_id=0):
+    cdef int boxes_num = dets.shape[0]
+    cdef int boxes_dim = dets.shape[1]
+    cdef int num_out
+    cdef np.ndarray[np.int32_t, ndim=1] \
+        keep = np.zeros(boxes_num, dtype=np.int32)
+    cdef np.ndarray[np.float32_t, ndim=1] \
+        scores = dets[:, 4]
+    cdef np.ndarray[np.int_t, ndim=1] \
+        order = scores.argsort()[::-1]
+    cdef np.ndarray[np.float32_t, ndim=2] \
+        sorted_dets = dets[order, :]
+    _nms(&keep[0], &num_out, &sorted_dets[0, 0], boxes_num, boxes_dim, thresh, device_id)
+    keep = keep[:num_out]
+    return list(order[keep])
diff --git a/lib/nms/nms_kernel.cu b/lib/nms/nms_kernel.cu
new file mode 100644
index 0000000..038a590
--- /dev/null
+++ b/lib/nms/nms_kernel.cu
@@ -0,0 +1,144 @@
+// ------------------------------------------------------------------
+// Faster R-CNN
+// Copyright (c) 2015 Microsoft
+// Licensed under The MIT License [see fast-rcnn/LICENSE for details]
+// Written by Shaoqing Ren
+// ------------------------------------------------------------------
+
+#include "gpu_nms.hpp"
+#include <vector>
+#include <iostream>
+
+#define CUDA_CHECK(condition) \
+  /* Code block avoids redefinition of cudaError_t error */ \
+  do { \
+    cudaError_t error = condition; \
+    if (error != cudaSuccess) { \
+      std::cout << cudaGetErrorString(error) << std::endl; \
+    } \
+  } while (0)
+
+#define DIVUP(m,n) ((m) / (n) + ((m) % (n) > 0))
+int const threadsPerBlock = sizeof(unsigned long long) * 8;
+
+__device__ inline float devIoU(float const * const a, float const * const b) {
+  float left = max(a[0], b[0]), right = min(a[2], b[2]);
+  float top = max(a[1], b[1]), bottom = min(a[3], b[3]);
+  float width = max(right - left + 1, 0.f), height = max(bottom - top + 1, 0.f);
+  float interS = width * height;
+  float Sa = (a[2] - a[0] + 1) * (a[3] - a[1] + 1);
+  float Sb = (b[2] - b[0] + 1) * (b[3] - b[1] + 1);
+  return interS / (Sa + Sb - interS);
+}
+
+__global__ void nms_kernel(const int n_boxes, const float nms_overlap_thresh,
+                           const float *dev_boxes, unsigned long long *dev_mask) {
+  const int row_start = blockIdx.y;
+  const int col_start = blockIdx.x;
+
+  // if (row_start > col_start) return;
+
+  const int row_size =
+        min(n_boxes - row_start * threadsPerBlock, threadsPerBlock);
+  const int col_size =
+        min(n_boxes - col_start * threadsPerBlock, threadsPerBlock);
+
+  __shared__ float block_boxes[threadsPerBlock * 5];
+  if (threadIdx.x < col_size) {
+    block_boxes[threadIdx.x * 5 + 0] =
+        dev_boxes[(threadsPerBlock * col_start + threadIdx.x) * 5 + 0];
+    block_boxes[threadIdx.x * 5 + 1] =
+        dev_boxes[(threadsPerBlock * col_start + threadIdx.x) * 5 + 1];
+    block_boxes[threadIdx.x * 5 + 2] =
+        dev_boxes[(threadsPerBlock * col_start + threadIdx.x) * 5 + 2];
+    block_boxes[threadIdx.x * 5 + 3] =
+        dev_boxes[(threadsPerBlock * col_start + threadIdx.x) * 5 + 3];
+    block_boxes[threadIdx.x * 5 + 4] =
+        dev_boxes[(threadsPerBlock * col_start + threadIdx.x) * 5 + 4];
+  }
+  __syncthreads();
+
+  if (threadIdx.x < row_size) {
+    const int cur_box_idx = threadsPerBlock * row_start + threadIdx.x;
+    const float *cur_box = dev_boxes + cur_box_idx * 5;
+    int i = 0;
+    unsigned long long t = 0;
+    int start = 0;
+    if (row_start == col_start) {
+      start = threadIdx.x + 1;
+    }
+    for (i = start; i < col_size; i++) {
+      if (devIoU(cur_box, block_boxes + i * 5) > nms_overlap_thresh) {
+        t |= 1ULL << i;
+      }
+    }
+    const int col_blocks = DIVUP(n_boxes, threadsPerBlock);
+    dev_mask[cur_box_idx * col_blocks + col_start] = t;
+  }
+}
+
+void _set_device(int device_id) {
+  int current_device;
+  CUDA_CHECK(cudaGetDevice(&current_device));
+  if (current_device == device_id) {
+    return;
+  }
+  // The call to cudaSetDevice must come before any calls to Get, which
+  // may perform initialization using the GPU.
+  CUDA_CHECK(cudaSetDevice(device_id));
+}
+
+void _nms(int* keep_out, int* num_out, const float* boxes_host, int boxes_num,
+          int boxes_dim, float nms_overlap_thresh, int device_id) {
+  _set_device(device_id);
+
+  float* boxes_dev = NULL;
+  unsigned long long* mask_dev = NULL;
+
+  const int col_blocks = DIVUP(boxes_num, threadsPerBlock);
+
+  CUDA_CHECK(cudaMalloc(&boxes_dev,
+                        boxes_num * boxes_dim * sizeof(float)));
+  CUDA_CHECK(cudaMemcpy(boxes_dev,
+                        boxes_host,
+                        boxes_num * boxes_dim * sizeof(float),
+                        cudaMemcpyHostToDevice));
+
+  CUDA_CHECK(cudaMalloc(&mask_dev,
+                        boxes_num * col_blocks * sizeof(unsigned long long)));
+
+  dim3 blocks(DIVUP(boxes_num, threadsPerBlock),
+              DIVUP(boxes_num, threadsPerBlock));
+  dim3 threads(threadsPerBlock);
+  nms_kernel<<<blocks, threads>>>(boxes_num,
+                                  nms_overlap_thresh,
+                                  boxes_dev,
+                                  mask_dev);
+
+  std::vector<unsigned long long> mask_host(boxes_num * col_blocks);
+  CUDA_CHECK(cudaMemcpy(&mask_host[0],
+                        mask_dev,
+                        sizeof(unsigned long long) * boxes_num * col_blocks,
+                        cudaMemcpyDeviceToHost));
+
+  std::vector<unsigned long long> remv(col_blocks);
+  memset(&remv[0], 0, sizeof(unsigned long long) * col_blocks);
+
+  int num_to_keep = 0;
+  for (int i = 0; i < boxes_num; i++) {
+    int nblock = i / threadsPerBlock;
+    int inblock = i % threadsPerBlock;
+
+    if (!(remv[nblock] & (1ULL << inblock))) {
+      keep_out[num_to_keep++] = i;
+      unsigned long long *p = &mask_host[0] + i * col_blocks;
+      for (int j = nblock; j < col_blocks; j++) {
+        remv[j] |= p[j];
+      }
+    }
+  }
+  *num_out = num_to_keep;
+
+  CUDA_CHECK(cudaFree(boxes_dev));
+  CUDA_CHECK(cudaFree(mask_dev));
+}
diff --git a/lib/nms/py_cpu_nms.py b/lib/nms/py_cpu_nms.py
new file mode 100644
index 0000000..54e7b25
--- /dev/null
+++ b/lib/nms/py_cpu_nms.py
@@ -0,0 +1,38 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+import numpy as np
+
+def py_cpu_nms(dets, thresh):
+    """Pure Python NMS baseline."""
+    x1 = dets[:, 0]
+    y1 = dets[:, 1]
+    x2 = dets[:, 2]
+    y2 = dets[:, 3]
+    scores = dets[:, 4]
+
+    areas = (x2 - x1 + 1) * (y2 - y1 + 1)
+    order = scores.argsort()[::-1]
+
+    keep = []
+    while order.size > 0:
+        i = order[0]
+        keep.append(i)
+        xx1 = np.maximum(x1[i], x1[order[1:]])
+        yy1 = np.maximum(y1[i], y1[order[1:]])
+        xx2 = np.minimum(x2[i], x2[order[1:]])
+        yy2 = np.minimum(y2[i], y2[order[1:]])
+
+        w = np.maximum(0.0, xx2 - xx1 + 1)
+        h = np.maximum(0.0, yy2 - yy1 + 1)
+        inter = w * h
+        ovr = inter / (areas[i] + areas[order[1:]] - inter)
+
+        inds = np.where(ovr <= thresh)[0]
+        order = order[inds + 1]
+
+    return keep
diff --git a/lib/pycocotools/UPSTREAM_REV b/lib/pycocotools/UPSTREAM_REV
new file mode 100644
index 0000000..706219b
--- /dev/null
+++ b/lib/pycocotools/UPSTREAM_REV
@@ -0,0 +1 @@
+https://github.com/pdollar/coco/commit/3ac47c77ebd5a1ed4254a98b7fbf2ef4765a3574
diff --git a/lib/pycocotools/__init__.py b/lib/pycocotools/__init__.py
new file mode 100644
index 0000000..3f7d85b
--- /dev/null
+++ b/lib/pycocotools/__init__.py
@@ -0,0 +1 @@
+__author__ = 'tylin'
diff --git a/lib/pycocotools/_mask.pyx b/lib/pycocotools/_mask.pyx
new file mode 100644
index 0000000..e08f1f4
--- /dev/null
+++ b/lib/pycocotools/_mask.pyx
@@ -0,0 +1,291 @@
+# distutils: language = c
+# distutils: sources = ../MatlabAPI/private/maskApi.c
+
+#**************************************************************************
+# Microsoft COCO Toolbox.      version 2.0
+# Data, paper, and tutorials available at:  http://mscoco.org/
+# Code written by Piotr Dollar and Tsung-Yi Lin, 2015.
+# Licensed under the Simplified BSD License [see coco/license.txt]
+#**************************************************************************
+
+__author__ = 'tsungyi'
+
+# import both Python-level and C-level symbols of Numpy
+# the API uses Numpy to interface C and Python
+import numpy as np
+cimport numpy as np
+from libc.stdlib cimport malloc, free
+
+# intialized Numpy. must do.
+np.import_array()
+
+# import numpy C function
+# we use PyArray_ENABLEFLAGS to make Numpy ndarray responsible to memoery management
+cdef extern from "numpy/arrayobject.h":
+    void PyArray_ENABLEFLAGS(np.ndarray arr, int flags)
+
+# Declare the prototype of the C functions in MaskApi.h
+cdef extern from "maskApi.h":
+    ctypedef unsigned int uint
+    ctypedef unsigned long siz
+    ctypedef unsigned char byte
+    ctypedef double* BB
+    ctypedef struct RLE:
+        siz h,
+        siz w,
+        siz m,
+        uint* cnts,
+    void rlesInit( RLE **R, siz n )
+    void rleEncode( RLE *R, const byte *M, siz h, siz w, siz n )
+    void rleDecode( const RLE *R, byte *mask, siz n )
+    void rleMerge( const RLE *R, RLE *M, siz n, bint intersect )
+    void rleArea( const RLE *R, siz n, uint *a )
+    void rleIou( RLE *dt, RLE *gt, siz m, siz n, byte *iscrowd, double *o )
+    void bbIou( BB dt, BB gt, siz m, siz n, byte *iscrowd, double *o )
+    void rleToBbox( const RLE *R, BB bb, siz n )
+    void rleFrBbox( RLE *R, const BB bb, siz h, siz w, siz n )
+    void rleFrPoly( RLE *R, const double *xy, siz k, siz h, siz w )
+    char* rleToString( const RLE *R )
+    void rleFrString( RLE *R, char *s, siz h, siz w )
+
+# python class to wrap RLE array in C
+# the class handles the memory allocation and deallocation
+cdef class RLEs:
+    cdef RLE *_R
+    cdef siz _n
+
+    def __cinit__(self, siz n =0):
+        rlesInit(&self._R, n)
+        self._n = n
+
+    # free the RLE array here
+    def __dealloc__(self):
+        if self._R is not NULL:
+            for i in range(self._n):
+                free(self._R[i].cnts)
+            free(self._R)
+    def __getattr__(self, key):
+        if key == 'n':
+            return self._n
+        raise AttributeError(key)
+
+# python class to wrap Mask array in C
+# the class handles the memory allocation and deallocation
+cdef class Masks:
+    cdef byte *_mask
+    cdef siz _h
+    cdef siz _w
+    cdef siz _n
+
+    def __cinit__(self, h, w, n):
+        self._mask = <byte*> malloc(h*w*n* sizeof(byte))
+        self._h = h
+        self._w = w
+        self._n = n
+    # def __dealloc__(self):
+        # the memory management of _mask has been passed to np.ndarray
+        # it doesn't need to be freed here
+
+    # called when passing into np.array() and return an np.ndarray in column-major order
+    def __array__(self):
+        cdef np.npy_intp shape[1]
+        shape[0] = <np.npy_intp> self._h*self._w*self._n
+        # Create a 1D array, and reshape it to fortran/Matlab column-major array
+        ndarray = np.PyArray_SimpleNewFromData(1, shape, np.NPY_UINT8, self._mask).reshape((self._h, self._w, self._n), order='F')
+        # The _mask allocated by Masks is now handled by ndarray
+        PyArray_ENABLEFLAGS(ndarray, np.NPY_OWNDATA)
+        return ndarray
+
+# internal conversion from Python RLEs object to compressed RLE format
+def _toString(RLEs Rs):
+    cdef siz n = Rs.n
+    cdef bytes py_string
+    cdef char* c_string
+    objs = []
+    for i in range(n):
+        c_string = rleToString( <RLE*> &Rs._R[i] )
+        py_string = c_string
+        objs.append({
+            'size': [Rs._R[i].h, Rs._R[i].w],
+            'counts': py_string
+        })
+        free(c_string)
+    return objs
+
+# internal conversion from compressed RLE format to Python RLEs object
+def _frString(rleObjs):
+    cdef siz n = len(rleObjs)
+    Rs = RLEs(n)
+    cdef bytes py_string
+    cdef char* c_string
+    for i, obj in enumerate(rleObjs):
+        py_string = str(obj['counts'])
+        c_string = py_string
+        rleFrString( <RLE*> &Rs._R[i], <char*> c_string, obj['size'][0], obj['size'][1] )
+    return Rs
+
+# encode mask to RLEs objects
+# list of RLE string can be generated by RLEs member function
+def encode(np.ndarray[np.uint8_t, ndim=3, mode='fortran'] mask):
+    h, w, n = mask.shape[0], mask.shape[1], mask.shape[2]
+    cdef RLEs Rs = RLEs(n)
+    rleEncode(Rs._R,<byte*>mask.data,h,w,n)
+    objs = _toString(Rs)
+    return objs
+
+# decode mask from compressed list of RLE string or RLEs object
+def decode(rleObjs):
+    cdef RLEs Rs = _frString(rleObjs)
+    h, w, n = Rs._R[0].h, Rs._R[0].w, Rs._n
+    masks = Masks(h, w, n)
+    rleDecode( <RLE*>Rs._R, masks._mask, n );
+    return np.array(masks)
+
+def merge(rleObjs, bint intersect=0):
+    cdef RLEs Rs = _frString(rleObjs)
+    cdef RLEs R = RLEs(1)
+    rleMerge(<RLE*>Rs._R, <RLE*> R._R, <siz> Rs._n, intersect)
+    obj = _toString(R)[0]
+    return obj
+
+def area(rleObjs):
+    cdef RLEs Rs = _frString(rleObjs)
+    cdef uint* _a = <uint*> malloc(Rs._n* sizeof(uint))
+    rleArea(Rs._R, Rs._n, _a)
+    cdef np.npy_intp shape[1]
+    shape[0] = <np.npy_intp> Rs._n
+    a = np.array((Rs._n, ), dtype=np.uint8)
+    a = np.PyArray_SimpleNewFromData(1, shape, np.NPY_UINT32, _a)
+    PyArray_ENABLEFLAGS(a, np.NPY_OWNDATA)
+    return a
+
+# iou computation. support function overload (RLEs-RLEs and bbox-bbox).
+def iou( dt, gt, pyiscrowd ):
+    def _preproc(objs):
+        if len(objs) == 0:
+            return objs
+        if type(objs) == np.ndarray:
+            if len(objs.shape) == 1:
+                objs = objs.reshape((objs[0], 1))
+            # check if it's Nx4 bbox
+            if not len(objs.shape) == 2 or not objs.shape[1] == 4:
+                raise Exception('numpy ndarray input is only for *bounding boxes* and should have Nx4 dimension')
+            objs = objs.astype(np.double)
+        elif type(objs) == list:
+            # check if list is in box format and convert it to np.ndarray
+            isbox = np.all(np.array([(len(obj)==4) and ((type(obj)==list) or (type(obj)==np.ndarray)) for obj in objs]))
+            isrle = np.all(np.array([type(obj) == dict for obj in objs]))
+            if isbox:
+                objs = np.array(objs, dtype=np.double)
+                if len(objs.shape) == 1:
+                    objs = objs.reshape((1,objs.shape[0]))
+            elif isrle:
+                objs = _frString(objs)
+            else:
+                raise Exception('list input can be bounding box (Nx4) or RLEs ([RLE])')
+        else:
+            raise Exception('unrecognized type.  The following type: RLEs (rle), np.ndarray (box), and list (box) are supported.')
+        return objs
+    def _rleIou(RLEs dt, RLEs gt, np.ndarray[np.uint8_t, ndim=1] iscrowd, siz m, siz n, np.ndarray[np.double_t,  ndim=1] _iou):
+        rleIou( <RLE*> dt._R, <RLE*> gt._R, m, n, <byte*> iscrowd.data, <double*> _iou.data )
+    def _bbIou(np.ndarray[np.double_t, ndim=2] dt, np.ndarray[np.double_t, ndim=2] gt, np.ndarray[np.uint8_t, ndim=1] iscrowd, siz m, siz n, np.ndarray[np.double_t, ndim=1] _iou):
+        bbIou( <BB> dt.data, <BB> gt.data, m, n, <byte*> iscrowd.data, <double*>_iou.data )
+    def _len(obj):
+        cdef siz N = 0
+        if type(obj) == RLEs:
+            N = obj.n
+        elif len(obj)==0:
+            pass
+        elif type(obj) == np.ndarray:
+            N = obj.shape[0]
+        return N
+    # convert iscrowd to numpy array
+    cdef np.ndarray[np.uint8_t, ndim=1] iscrowd = np.array(pyiscrowd, dtype=np.uint8)
+    # simple type checking
+    cdef siz m, n
+    dt = _preproc(dt)
+    gt = _preproc(gt)
+    m = _len(dt)
+    n = _len(gt)
+    if m == 0 or n == 0:
+        return []
+    if not type(dt) == type(gt):
+        raise Exception('The dt and gt should have the same data type, either RLEs, list or np.ndarray')
+
+    # define local variables
+    cdef double* _iou = <double*> 0
+    cdef np.npy_intp shape[1]
+    # check type and assign iou function
+    if type(dt) == RLEs:
+        _iouFun = _rleIou
+    elif type(dt) == np.ndarray:
+        _iouFun = _bbIou
+    else:
+        raise Exception('input data type not allowed.')
+    _iou = <double*> malloc(m*n* sizeof(double))
+    iou = np.zeros((m*n, ), dtype=np.double)
+    shape[0] = <np.npy_intp> m*n
+    iou = np.PyArray_SimpleNewFromData(1, shape, np.NPY_DOUBLE, _iou)
+    PyArray_ENABLEFLAGS(iou, np.NPY_OWNDATA)
+    _iouFun(dt, gt, iscrowd, m, n, iou)
+    return iou.reshape((m,n), order='F')
+
+def toBbox( rleObjs ):
+    cdef RLEs Rs = _frString(rleObjs)
+    cdef siz n = Rs.n
+    cdef BB _bb = <BB> malloc(4*n* sizeof(double))
+    rleToBbox( <const RLE*> Rs._R, _bb, n )
+    cdef np.npy_intp shape[1]
+    shape[0] = <np.npy_intp> 4*n
+    bb = np.array((1,4*n), dtype=np.double)
+    bb = np.PyArray_SimpleNewFromData(1, shape, np.NPY_DOUBLE, _bb).reshape((n, 4))
+    PyArray_ENABLEFLAGS(bb, np.NPY_OWNDATA)
+    return bb
+
+def frBbox(np.ndarray[np.double_t, ndim=2] bb, siz h, siz w ):
+    cdef siz n = bb.shape[0]
+    Rs = RLEs(n)
+    rleFrBbox( <RLE*> Rs._R, <const BB> bb.data, h, w, n )
+    objs = _toString(Rs)
+    return objs
+
+def frPoly( poly, siz h, siz w ):
+    cdef np.ndarray[np.double_t, ndim=1] np_poly
+    n = len(poly)
+    Rs = RLEs(n)
+    for i, p in enumerate(poly):
+        np_poly = np.array(p, dtype=np.double, order='F')
+        rleFrPoly( <RLE*>&Rs._R[i], <const double*> np_poly.data, len(np_poly)/2, h, w )
+    objs = _toString(Rs)
+    return objs
+
+def frUncompressedRLE(ucRles, siz h, siz w):
+    cdef np.ndarray[np.uint32_t, ndim=1] cnts
+    cdef RLE R
+    cdef uint *data
+    n = len(ucRles)
+    objs = []
+    for i in range(n):
+        Rs = RLEs(1)
+        cnts = np.array(ucRles[i]['counts'], dtype=np.uint32)
+        # time for malloc can be saved here but it's fine
+        data = <uint*> malloc(len(cnts)* sizeof(uint))
+        for j in range(len(cnts)):
+            data[j] = <uint> cnts[j]
+        R = RLE(ucRles[i]['size'][0], ucRles[i]['size'][1], len(cnts), <uint*> data)
+        Rs._R[0] = R
+        objs.append(_toString(Rs)[0])
+    return objs
+
+def frPyObjects(pyobj, siz h, w):
+    if type(pyobj) == np.ndarray:
+        objs = frBbox(pyobj, h, w )
+    elif type(pyobj) == list and len(pyobj[0]) == 4:
+        objs = frBbox(pyobj, h, w )
+    elif type(pyobj) == list and len(pyobj[0]) > 4:
+        objs = frPoly(pyobj, h, w )
+    elif type(pyobj) == list and type(pyobj[0]) == dict:
+        objs = frUncompressedRLE(pyobj, h, w)
+    else:
+        raise Exception('input type is not supported.')
+    return objs
diff --git a/lib/pycocotools/coco.py b/lib/pycocotools/coco.py
new file mode 100644
index 0000000..5d9f6b8
--- /dev/null
+++ b/lib/pycocotools/coco.py
@@ -0,0 +1,351 @@
+__author__ = 'tylin'
+__version__ = '1.0.1'
+# Interface for accessing the Microsoft COCO dataset.
+
+# Microsoft COCO is a large image dataset designed for object detection,
+# segmentation, and caption generation. pycocotools is a Python API that
+# assists in loading, parsing and visualizing the annotations in COCO.
+# Please visit http://mscoco.org/ for more information on COCO, including
+# for the data, paper, and tutorials. The exact format of the annotations
+# is also described on the COCO website. For example usage of the pycocotools
+# please see pycocotools_demo.ipynb. In addition to this API, please download both
+# the COCO images and annotations in order to run the demo.
+
+# An alternative to using the API is to load the annotations directly
+# into Python dictionary
+# Using the API provides additional utility functions. Note that this API
+# supports both *instance* and *caption* annotations. In the case of
+# captions not all functions are defined (e.g. categories are undefined).
+
+# The following API functions are defined:
+#  COCO       - COCO api class that loads COCO annotation file and prepare data structures.
+#  decodeMask - Decode binary mask M encoded via run-length encoding.
+#  encodeMask - Encode binary mask M using run-length encoding.
+#  getAnnIds  - Get ann ids that satisfy given filter conditions.
+#  getCatIds  - Get cat ids that satisfy given filter conditions.
+#  getImgIds  - Get img ids that satisfy given filter conditions.
+#  loadAnns   - Load anns with the specified ids.
+#  loadCats   - Load cats with the specified ids.
+#  loadImgs   - Load imgs with the specified ids.
+#  segToMask  - Convert polygon segmentation to binary mask.
+#  showAnns   - Display the specified annotations.
+#  loadRes    - Load algorithm results and create API for accessing them.
+#  download   - Download COCO images from mscoco.org server.
+# Throughout the API "ann"=annotation, "cat"=category, and "img"=image.
+# Help on each functions can be accessed by: "help COCO>function".
+
+# See also COCO>decodeMask,
+# COCO>encodeMask, COCO>getAnnIds, COCO>getCatIds,
+# COCO>getImgIds, COCO>loadAnns, COCO>loadCats,
+# COCO>loadImgs, COCO>segToMask, COCO>showAnns
+
+# Microsoft COCO Toolbox.      version 2.0
+# Data, paper, and tutorials available at:  http://mscoco.org/
+# Code written by Piotr Dollar and Tsung-Yi Lin, 2014.
+# Licensed under the Simplified BSD License [see bsd.txt]
+
+import json
+import datetime
+import time
+import matplotlib.pyplot as plt
+from matplotlib.collections import PatchCollection
+from matplotlib.patches import Polygon
+import numpy as np
+from skimage.draw import polygon
+import urllib
+import copy
+import itertools
+import mask
+import os
+
+class COCO:
+    def __init__(self, annotation_file=None):
+        """
+        Constructor of Microsoft COCO helper class for reading and visualizing annotations.
+        :param annotation_file (str): location of annotation file
+        :param image_folder (str): location to the folder that hosts images.
+        :return:
+        """
+        # load dataset
+        self.dataset = {}
+        self.anns = []
+        self.imgToAnns = {}
+        self.catToImgs = {}
+        self.imgs = {}
+        self.cats = {}
+        if not annotation_file == None:
+            print 'loading annotations into memory...'
+            tic = time.time()
+            dataset = json.load(open(annotation_file, 'r'))
+            print 'Done (t=%0.2fs)'%(time.time()- tic)
+            self.dataset = dataset
+            self.createIndex()
+
+    def createIndex(self):
+        # create index
+        print 'creating index...'
+        anns = {}
+        imgToAnns = {}
+        catToImgs = {}
+        cats = {}
+        imgs = {}
+        if 'annotations' in self.dataset:
+            imgToAnns = {ann['image_id']: [] for ann in self.dataset['annotations']}
+            anns =      {ann['id']:       [] for ann in self.dataset['annotations']}
+            for ann in self.dataset['annotations']:
+                imgToAnns[ann['image_id']] += [ann]
+                anns[ann['id']] = ann
+
+        if 'images' in self.dataset:
+            imgs      = {im['id']: {} for im in self.dataset['images']}
+            for img in self.dataset['images']:
+                imgs[img['id']] = img
+
+        if 'categories' in self.dataset:
+            cats = {cat['id']: [] for cat in self.dataset['categories']}
+            for cat in self.dataset['categories']:
+                cats[cat['id']] = cat
+            catToImgs = {cat['id']: [] for cat in self.dataset['categories']}
+            if 'annotations' in self.dataset:
+                for ann in self.dataset['annotations']:
+                    catToImgs[ann['category_id']] += [ann['image_id']]
+
+        print 'index created!'
+
+        # create class members
+        self.anns = anns
+        self.imgToAnns = imgToAnns
+        self.catToImgs = catToImgs
+        self.imgs = imgs
+        self.cats = cats
+
+    def info(self):
+        """
+        Print information about the annotation file.
+        :return:
+        """
+        for key, value in self.dataset['info'].items():
+            print '%s: %s'%(key, value)
+
+    def getAnnIds(self, imgIds=[], catIds=[], areaRng=[], iscrowd=None):
+        """
+        Get ann ids that satisfy given filter conditions. default skips that filter
+        :param imgIds  (int array)     : get anns for given imgs
+               catIds  (int array)     : get anns for given cats
+               areaRng (float array)   : get anns for given area range (e.g. [0 inf])
+               iscrowd (boolean)       : get anns for given crowd label (False or True)
+        :return: ids (int array)       : integer array of ann ids
+        """
+        imgIds = imgIds if type(imgIds) == list else [imgIds]
+        catIds = catIds if type(catIds) == list else [catIds]
+
+        if len(imgIds) == len(catIds) == len(areaRng) == 0:
+            anns = self.dataset['annotations']
+        else:
+            if not len(imgIds) == 0:
+                # this can be changed by defaultdict
+                lists = [self.imgToAnns[imgId] for imgId in imgIds if imgId in self.imgToAnns]
+                anns = list(itertools.chain.from_iterable(lists))
+            else:
+                anns = self.dataset['annotations']
+            anns = anns if len(catIds)  == 0 else [ann for ann in anns if ann['category_id'] in catIds]
+            anns = anns if len(areaRng) == 0 else [ann for ann in anns if ann['area'] > areaRng[0] and ann['area'] < areaRng[1]]
+        if not iscrowd == None:
+            ids = [ann['id'] for ann in anns if ann['iscrowd'] == iscrowd]
+        else:
+            ids = [ann['id'] for ann in anns]
+        return ids
+
+    def getCatIds(self, catNms=[], supNms=[], catIds=[]):
+        """
+        filtering parameters. default skips that filter.
+        :param catNms (str array)  : get cats for given cat names
+        :param supNms (str array)  : get cats for given supercategory names
+        :param catIds (int array)  : get cats for given cat ids
+        :return: ids (int array)   : integer array of cat ids
+        """
+        catNms = catNms if type(catNms) == list else [catNms]
+        supNms = supNms if type(supNms) == list else [supNms]
+        catIds = catIds if type(catIds) == list else [catIds]
+
+        if len(catNms) == len(supNms) == len(catIds) == 0:
+            cats = self.dataset['categories']
+        else:
+            cats = self.dataset['categories']
+            cats = cats if len(catNms) == 0 else [cat for cat in cats if cat['name']          in catNms]
+            cats = cats if len(supNms) == 0 else [cat for cat in cats if cat['supercategory'] in supNms]
+            cats = cats if len(catIds) == 0 else [cat for cat in cats if cat['id']            in catIds]
+        ids = [cat['id'] for cat in cats]
+        return ids
+
+    def getImgIds(self, imgIds=[], catIds=[]):
+        '''
+        Get img ids that satisfy given filter conditions.
+        :param imgIds (int array) : get imgs for given ids
+        :param catIds (int array) : get imgs with all given cats
+        :return: ids (int array)  : integer array of img ids
+        '''
+        imgIds = imgIds if type(imgIds) == list else [imgIds]
+        catIds = catIds if type(catIds) == list else [catIds]
+
+        if len(imgIds) == len(catIds) == 0:
+            ids = self.imgs.keys()
+        else:
+            ids = set(imgIds)
+            for i, catId in enumerate(catIds):
+                if i == 0 and len(ids) == 0:
+                    ids = set(self.catToImgs[catId])
+                else:
+                    ids &= set(self.catToImgs[catId])
+        return list(ids)
+
+    def loadAnns(self, ids=[]):
+        """
+        Load anns with the specified ids.
+        :param ids (int array)       : integer ids specifying anns
+        :return: anns (object array) : loaded ann objects
+        """
+        if type(ids) == list:
+            return [self.anns[id] for id in ids]
+        elif type(ids) == int:
+            return [self.anns[ids]]
+
+    def loadCats(self, ids=[]):
+        """
+        Load cats with the specified ids.
+        :param ids (int array)       : integer ids specifying cats
+        :return: cats (object array) : loaded cat objects
+        """
+        if type(ids) == list:
+            return [self.cats[id] for id in ids]
+        elif type(ids) == int:
+            return [self.cats[ids]]
+
+    def loadImgs(self, ids=[]):
+        """
+        Load anns with the specified ids.
+        :param ids (int array)       : integer ids specifying img
+        :return: imgs (object array) : loaded img objects
+        """
+        if type(ids) == list:
+            return [self.imgs[id] for id in ids]
+        elif type(ids) == int:
+            return [self.imgs[ids]]
+
+    def showAnns(self, anns):
+        """
+        Display the specified annotations.
+        :param anns (array of object): annotations to display
+        :return: None
+        """
+        if len(anns) == 0:
+            return 0
+        if 'segmentation' in anns[0]:
+            datasetType = 'instances'
+        elif 'caption' in anns[0]:
+            datasetType = 'captions'
+        if datasetType == 'instances':
+            ax = plt.gca()
+            polygons = []
+            color = []
+            for ann in anns:
+                c = np.random.random((1, 3)).tolist()[0]
+                if type(ann['segmentation']) == list:
+                    # polygon
+                    for seg in ann['segmentation']:
+                        poly = np.array(seg).reshape((len(seg)/2, 2))
+                        polygons.append(Polygon(poly, True,alpha=0.4))
+                        color.append(c)
+                else:
+                    # mask
+                    t = self.imgs[ann['image_id']]
+                    if type(ann['segmentation']['counts']) == list:
+                        rle = mask.frPyObjects([ann['segmentation']], t['height'], t['width'])
+                    else:
+                        rle = [ann['segmentation']]
+                    m = mask.decode(rle)
+                    img = np.ones( (m.shape[0], m.shape[1], 3) )
+                    if ann['iscrowd'] == 1:
+                        color_mask = np.array([2.0,166.0,101.0])/255
+                    if ann['iscrowd'] == 0:
+                        color_mask = np.random.random((1, 3)).tolist()[0]
+                    for i in range(3):
+                        img[:,:,i] = color_mask[i]
+                    ax.imshow(np.dstack( (img, m*0.5) ))
+            p = PatchCollection(polygons, facecolors=color, edgecolors=(0,0,0,1), linewidths=3, alpha=0.4)
+            ax.add_collection(p)
+        elif datasetType == 'captions':
+            for ann in anns:
+                print ann['caption']
+
+    def loadRes(self, resFile):
+        """
+        Load result file and return a result api object.
+        :param   resFile (str)     : file name of result file
+        :return: res (obj)         : result api object
+        """
+        res = COCO()
+        res.dataset['images'] = [img for img in self.dataset['images']]
+        # res.dataset['info'] = copy.deepcopy(self.dataset['info'])
+        # res.dataset['licenses'] = copy.deepcopy(self.dataset['licenses'])
+
+        print 'Loading and preparing results...     '
+        tic = time.time()
+        anns    = json.load(open(resFile))
+        assert type(anns) == list, 'results in not an array of objects'
+        annsImgIds = [ann['image_id'] for ann in anns]
+        assert set(annsImgIds) == (set(annsImgIds) & set(self.getImgIds())), \
+               'Results do not correspond to current coco set'
+        if 'caption' in anns[0]:
+            imgIds = set([img['id'] for img in res.dataset['images']]) & set([ann['image_id'] for ann in anns])
+            res.dataset['images'] = [img for img in res.dataset['images'] if img['id'] in imgIds]
+            for id, ann in enumerate(anns):
+                ann['id'] = id+1
+        elif 'bbox' in anns[0] and not anns[0]['bbox'] == []:
+            res.dataset['categories'] = copy.deepcopy(self.dataset['categories'])
+            for id, ann in enumerate(anns):
+                bb = ann['bbox']
+                x1, x2, y1, y2 = [bb[0], bb[0]+bb[2], bb[1], bb[1]+bb[3]]
+                if not 'segmentation' in ann:
+                    ann['segmentation'] = [[x1, y1, x1, y2, x2, y2, x2, y1]]
+                ann['area'] = bb[2]*bb[3]
+                ann['id'] = id+1
+                ann['iscrowd'] = 0
+        elif 'segmentation' in anns[0]:
+            res.dataset['categories'] = copy.deepcopy(self.dataset['categories'])
+            for id, ann in enumerate(anns):
+                # now only support compressed RLE format as segmentation results
+                ann['area'] = mask.area([ann['segmentation']])[0]
+                if not 'bbox' in ann:
+                    ann['bbox'] = mask.toBbox([ann['segmentation']])[0]
+                ann['id'] = id+1
+                ann['iscrowd'] = 0
+        print 'DONE (t=%0.2fs)'%(time.time()- tic)
+
+        res.dataset['annotations'] = anns
+        res.createIndex()
+        return res
+
+    def download( self, tarDir = None, imgIds = [] ):
+        '''
+        Download COCO images from mscoco.org server.
+        :param tarDir (str): COCO results directory name
+               imgIds (list): images to be downloaded
+        :return:
+        '''
+        if tarDir is None:
+            print 'Please specify target directory'
+            return -1
+        if len(imgIds) == 0:
+            imgs = self.imgs.values()
+        else:
+            imgs = self.loadImgs(imgIds)
+        N = len(imgs)
+        if not os.path.exists(tarDir):
+            os.makedirs(tarDir)
+        for i, img in enumerate(imgs):
+            tic = time.time()
+            fname = os.path.join(tarDir, img['file_name'])
+            if not os.path.exists(fname):
+                urllib.urlretrieve(img['coco_url'], fname)
+            print 'downloaded %d/%d images (t=%.1fs)'%(i, N, time.time()- tic)
diff --git a/lib/pycocotools/cocoeval.py b/lib/pycocotools/cocoeval.py
new file mode 100644
index 0000000..f389eb0
--- /dev/null
+++ b/lib/pycocotools/cocoeval.py
@@ -0,0 +1,444 @@
+__author__ = 'tsungyi'
+
+import numpy as np
+import datetime
+import time
+from collections import defaultdict
+import mask
+import copy
+
+class COCOeval:
+    # Interface for evaluating detection on the Microsoft COCO dataset.
+    #
+    # The usage for CocoEval is as follows:
+    #  cocoGt=..., cocoDt=...       # load dataset and results
+    #  E = CocoEval(cocoGt,cocoDt); # initialize CocoEval object
+    #  E.params.recThrs = ...;      # set parameters as desired
+    #  E.evaluate();                # run per image evaluation
+    #  E.accumulate();              # accumulate per image results
+    #  E.summarize();               # display summary metrics of results
+    # For example usage see evalDemo.m and http://mscoco.org/.
+    #
+    # The evaluation parameters are as follows (defaults in brackets):
+    #  imgIds     - [all] N img ids to use for evaluation
+    #  catIds     - [all] K cat ids to use for evaluation
+    #  iouThrs    - [.5:.05:.95] T=10 IoU thresholds for evaluation
+    #  recThrs    - [0:.01:1] R=101 recall thresholds for evaluation
+    #  areaRng    - [...] A=4 object area ranges for evaluation
+    #  maxDets    - [1 10 100] M=3 thresholds on max detections per image
+    #  useSegm    - [1] if true evaluate against ground-truth segments
+    #  useCats    - [1] if true use category labels for evaluation    # Note: if useSegm=0 the evaluation is run on bounding boxes.
+    # Note: if useCats=0 category labels are ignored as in proposal scoring.
+    # Note: multiple areaRngs [Ax2] and maxDets [Mx1] can be specified.
+    #
+    # evaluate(): evaluates detections on every image and every category and
+    # concats the results into the "evalImgs" with fields:
+    #  dtIds      - [1xD] id for each of the D detections (dt)
+    #  gtIds      - [1xG] id for each of the G ground truths (gt)
+    #  dtMatches  - [TxD] matching gt id at each IoU or 0
+    #  gtMatches  - [TxG] matching dt id at each IoU or 0
+    #  dtScores   - [1xD] confidence of each dt
+    #  gtIgnore   - [1xG] ignore flag for each gt
+    #  dtIgnore   - [TxD] ignore flag for each dt at each IoU
+    #
+    # accumulate(): accumulates the per-image, per-category evaluation
+    # results in "evalImgs" into the dictionary "eval" with fields:
+    #  params     - parameters used for evaluation
+    #  date       - date evaluation was performed
+    #  counts     - [T,R,K,A,M] parameter dimensions (see above)
+    #  precision  - [TxRxKxAxM] precision for every evaluation setting
+    #  recall     - [TxKxAxM] max recall for every evaluation setting
+    # Note: precision and recall==-1 for settings with no gt objects.
+    #
+    # See also coco, mask, pycocoDemo, pycocoEvalDemo
+    #
+    # Microsoft COCO Toolbox.      version 2.0
+    # Data, paper, and tutorials available at:  http://mscoco.org/
+    # Code written by Piotr Dollar and Tsung-Yi Lin, 2015.
+    # Licensed under the Simplified BSD License [see coco/license.txt]
+    def __init__(self, cocoGt=None, cocoDt=None):
+        '''
+        Initialize CocoEval using coco APIs for gt and dt
+        :param cocoGt: coco object with ground truth annotations
+        :param cocoDt: coco object with detection results
+        :return: None
+        '''
+        self.cocoGt   = cocoGt              # ground truth COCO API
+        self.cocoDt   = cocoDt              # detections COCO API
+        self.params   = {}                  # evaluation parameters
+        self.evalImgs = defaultdict(list)   # per-image per-category evaluation results [KxAxI] elements
+        self.eval     = {}                  # accumulated evaluation results
+        self._gts = defaultdict(list)       # gt for evaluation
+        self._dts = defaultdict(list)       # dt for evaluation
+        self.params = Params()              # parameters
+        self._paramsEval = {}               # parameters for evaluation
+        self.stats = []                     # result summarization
+        self.ious = {}                      # ious between all gts and dts
+        if not cocoGt is None:
+            self.params.imgIds = sorted(cocoGt.getImgIds())
+            self.params.catIds = sorted(cocoGt.getCatIds())
+
+
+    def _prepare(self):
+        '''
+        Prepare ._gts and ._dts for evaluation based on params
+        :return: None
+        '''
+        #
+        def _toMask(objs, coco):
+            # modify segmentation by reference
+            for obj in objs:
+                t = coco.imgs[obj['image_id']]
+                if type(obj['segmentation']) == list:
+                    if type(obj['segmentation'][0]) == dict:
+                        print 'debug'
+                    obj['segmentation'] = mask.frPyObjects(obj['segmentation'],t['height'],t['width'])
+                    if len(obj['segmentation']) == 1:
+                        obj['segmentation'] = obj['segmentation'][0]
+                    else:
+                        # an object can have multiple polygon regions
+                        # merge them into one RLE mask
+                        obj['segmentation'] = mask.merge(obj['segmentation'])
+                elif type(obj['segmentation']) == dict and type(obj['segmentation']['counts']) == list:
+                    obj['segmentation'] = mask.frPyObjects([obj['segmentation']],t['height'],t['width'])[0]
+                elif type(obj['segmentation']) == dict and \
+                     type(obj['segmentation']['counts'] == unicode or type(obj['segmentation']['counts']) == str):
+                    pass
+                else:
+                    raise Exception('segmentation format not supported.')
+        p = self.params
+        if p.useCats:
+            gts=self.cocoGt.loadAnns(self.cocoGt.getAnnIds(imgIds=p.imgIds, catIds=p.catIds))
+            dts=self.cocoDt.loadAnns(self.cocoDt.getAnnIds(imgIds=p.imgIds, catIds=p.catIds))
+        else:
+            gts=self.cocoGt.loadAnns(self.cocoGt.getAnnIds(imgIds=p.imgIds))
+            dts=self.cocoDt.loadAnns(self.cocoDt.getAnnIds(imgIds=p.imgIds))
+
+        if p.useSegm:
+            _toMask(gts, self.cocoGt)
+            _toMask(dts, self.cocoDt)
+        self._gts = defaultdict(list)       # gt for evaluation
+        self._dts = defaultdict(list)       # dt for evaluation
+        for gt in gts:
+            self._gts[gt['image_id'], gt['category_id']].append(gt)
+        for dt in dts:
+            self._dts[dt['image_id'], dt['category_id']].append(dt)
+        self.evalImgs = defaultdict(list)   # per-image per-category evaluation results
+        self.eval     = {}                  # accumulated evaluation results
+
+    def evaluate(self):
+        '''
+        Run per image evaluation on given images and store results (a list of dict) in self.evalImgs
+        :return: None
+        '''
+        tic = time.time()
+        print 'Running per image evaluation...      '
+        p = self.params
+        p.imgIds = list(np.unique(p.imgIds))
+        if p.useCats:
+            p.catIds = list(np.unique(p.catIds))
+        p.maxDets = sorted(p.maxDets)
+        self.params=p
+
+        self._prepare()
+        # loop through images, area range, max detection number
+        catIds = p.catIds if p.useCats else [-1]
+
+        computeIoU = self.computeIoU
+        self.ious = {(imgId, catId): computeIoU(imgId, catId) \
+                        for imgId in p.imgIds
+                        for catId in catIds}
+
+        evaluateImg = self.evaluateImg
+        maxDet = p.maxDets[-1]
+        self.evalImgs = [evaluateImg(imgId, catId, areaRng, maxDet)
+                 for catId in catIds
+                 for areaRng in p.areaRng
+                 for imgId in p.imgIds
+             ]
+        self._paramsEval = copy.deepcopy(self.params)
+        toc = time.time()
+        print 'DONE (t=%0.2fs).'%(toc-tic)
+
+    def computeIoU(self, imgId, catId):
+        p = self.params
+        if p.useCats:
+            gt = self._gts[imgId,catId]
+            dt = self._dts[imgId,catId]
+        else:
+            gt = [_ for cId in p.catIds for _ in self._gts[imgId,cId]]
+            dt = [_ for cId in p.catIds for _ in self._dts[imgId,cId]]
+        if len(gt) == 0 and len(dt) ==0:
+            return []
+        dt = sorted(dt, key=lambda x: -x['score'])
+        if len(dt) > p.maxDets[-1]:
+            dt=dt[0:p.maxDets[-1]]
+
+        if p.useSegm:
+            g = [g['segmentation'] for g in gt]
+            d = [d['segmentation'] for d in dt]
+        else:
+            g = [g['bbox'] for g in gt]
+            d = [d['bbox'] for d in dt]
+
+        # compute iou between each dt and gt region
+        iscrowd = [int(o['iscrowd']) for o in gt]
+        ious = mask.iou(d,g,iscrowd)
+        return ious
+
+    def evaluateImg(self, imgId, catId, aRng, maxDet):
+        '''
+        perform evaluation for single category and image
+        :return: dict (single image results)
+        '''
+        #
+        p = self.params
+        if p.useCats:
+            gt = self._gts[imgId,catId]
+            dt = self._dts[imgId,catId]
+        else:
+            gt = [_ for cId in p.catIds for _ in self._gts[imgId,cId]]
+            dt = [_ for cId in p.catIds for _ in self._dts[imgId,cId]]
+        if len(gt) == 0 and len(dt) ==0:
+            return None
+
+        for g in gt:
+            if 'ignore' not in g:
+                g['ignore'] = 0
+            if g['iscrowd'] == 1 or g['ignore'] or (g['area']<aRng[0] or g['area']>aRng[1]):
+                g['_ignore'] = 1
+            else:
+                g['_ignore'] = 0
+
+        # sort dt highest score first, sort gt ignore last
+        # gt = sorted(gt, key=lambda x: x['_ignore'])
+        gtind = [ind for (ind, g) in sorted(enumerate(gt), key=lambda (ind, g): g['_ignore']) ]
+
+        gt = [gt[ind] for ind in gtind]
+        dt = sorted(dt, key=lambda x: -x['score'])[0:maxDet]
+        iscrowd = [int(o['iscrowd']) for o in gt]
+        # load computed ious
+        N_iou = len(self.ious[imgId, catId])
+        ious = self.ious[imgId, catId][0:maxDet, np.array(gtind)] if N_iou >0 else self.ious[imgId, catId]
+
+        T = len(p.iouThrs)
+        G = len(gt)
+        D = len(dt)
+        gtm  = np.zeros((T,G))
+        dtm  = np.zeros((T,D))
+        gtIg = np.array([g['_ignore'] for g in gt])
+        dtIg = np.zeros((T,D))
+        if not len(ious)==0:
+            for tind, t in enumerate(p.iouThrs):
+                for dind, d in enumerate(dt):
+                    # information about best match so far (m=-1 -> unmatched)
+                    iou = min([t,1-1e-10])
+                    m   = -1
+                    for gind, g in enumerate(gt):
+                        # if this gt already matched, and not a crowd, continue
+                        if gtm[tind,gind]>0 and not iscrowd[gind]:
+                            continue
+                        # if dt matched to reg gt, and on ignore gt, stop
+                        if m>-1 and gtIg[m]==0 and gtIg[gind]==1:
+                            break
+                        # continue to next gt unless better match made
+                        if ious[dind,gind] < iou:
+                            continue
+                        # match successful and best so far, store appropriately
+                        iou=ious[dind,gind]
+                        m=gind
+                    # if match made store id of match for both dt and gt
+                    if m ==-1:
+                        continue
+                    dtIg[tind,dind] = gtIg[m]
+                    dtm[tind,dind]  = gt[m]['id']
+                    gtm[tind,m]     = d['id']
+        # set unmatched detections outside of area range to ignore
+        a = np.array([d['area']<aRng[0] or d['area']>aRng[1] for d in dt]).reshape((1, len(dt)))
+        dtIg = np.logical_or(dtIg, np.logical_and(dtm==0, np.repeat(a,T,0)))
+        # store results for given image and category
+        return {
+                'image_id':     imgId,
+                'category_id':  catId,
+                'aRng':         aRng,
+                'maxDet':       maxDet,
+                'dtIds':        [d['id'] for d in dt],
+                'gtIds':        [g['id'] for g in gt],
+                'dtMatches':    dtm,
+                'gtMatches':    gtm,
+                'dtScores':     [d['score'] for d in dt],
+                'gtIgnore':     gtIg,
+                'dtIgnore':     dtIg,
+            }
+
+    def accumulate(self, p = None):
+        '''
+        Accumulate per image evaluation results and store the result in self.eval
+        :param p: input params for evaluation
+        :return: None
+        '''
+        print 'Accumulating evaluation results...   '
+        tic = time.time()
+        if not self.evalImgs:
+            print 'Please run evaluate() first'
+        # allows input customized parameters
+        if p is None:
+            p = self.params
+        p.catIds = p.catIds if p.useCats == 1 else [-1]
+        T           = len(p.iouThrs)
+        R           = len(p.recThrs)
+        K           = len(p.catIds) if p.useCats else 1
+        A           = len(p.areaRng)
+        M           = len(p.maxDets)
+        precision   = -np.ones((T,R,K,A,M)) # -1 for the precision of absent categories
+        recall      = -np.ones((T,K,A,M))
+
+        # create dictionary for future indexing
+        _pe = self._paramsEval
+        catIds = _pe.catIds if _pe.useCats else [-1]
+        setK = set(catIds)
+        setA = set(map(tuple, _pe.areaRng))
+        setM = set(_pe.maxDets)
+        setI = set(_pe.imgIds)
+        # get inds to evaluate
+        k_list = [n for n, k in enumerate(p.catIds)  if k in setK]
+        m_list = [m for n, m in enumerate(p.maxDets) if m in setM]
+        a_list = [n for n, a in enumerate(map(lambda x: tuple(x), p.areaRng)) if a in setA]
+        i_list = [n for n, i in enumerate(p.imgIds)  if i in setI]
+        # K0 = len(_pe.catIds)
+        I0 = len(_pe.imgIds)
+        A0 = len(_pe.areaRng)
+        # retrieve E at each category, area range, and max number of detections
+        for k, k0 in enumerate(k_list):
+            Nk = k0*A0*I0
+            for a, a0 in enumerate(a_list):
+                Na = a0*I0
+                for m, maxDet in enumerate(m_list):
+                    E = [self.evalImgs[Nk+Na+i] for i in i_list]
+                    E = filter(None, E)
+                    if len(E) == 0:
+                        continue
+                    dtScores = np.concatenate([e['dtScores'][0:maxDet] for e in E])
+
+                    # different sorting method generates slightly different results.
+                    # mergesort is used to be consistent as Matlab implementation.
+                    inds = np.argsort(-dtScores, kind='mergesort')
+
+                    dtm  = np.concatenate([e['dtMatches'][:,0:maxDet] for e in E], axis=1)[:,inds]
+                    dtIg = np.concatenate([e['dtIgnore'][:,0:maxDet]  for e in E], axis=1)[:,inds]
+                    gtIg = np.concatenate([e['gtIgnore']  for e in E])
+                    npig = len([ig for ig in gtIg if ig == 0])
+                    if npig == 0:
+                        continue
+                    tps = np.logical_and(               dtm,  np.logical_not(dtIg) )
+                    fps = np.logical_and(np.logical_not(dtm), np.logical_not(dtIg) )
+
+                    tp_sum = np.cumsum(tps, axis=1).astype(dtype=np.float)
+                    fp_sum = np.cumsum(fps, axis=1).astype(dtype=np.float)
+                    for t, (tp, fp) in enumerate(zip(tp_sum, fp_sum)):
+                        tp = np.array(tp)
+                        fp = np.array(fp)
+                        nd = len(tp)
+                        rc = tp / npig
+                        pr = tp / (fp+tp+np.spacing(1))
+                        q  = np.zeros((R,))
+
+                        if nd:
+                            recall[t,k,a,m] = rc[-1]
+                        else:
+                            recall[t,k,a,m] = 0
+
+                        # numpy is slow without cython optimization for accessing elements
+                        # use python array gets significant speed improvement
+                        pr = pr.tolist(); q = q.tolist()
+
+                        for i in range(nd-1, 0, -1):
+                            if pr[i] > pr[i-1]:
+                                pr[i-1] = pr[i]
+
+                        inds = np.searchsorted(rc, p.recThrs)
+                        try:
+                            for ri, pi in enumerate(inds):
+                                q[ri] = pr[pi]
+                        except:
+                            pass
+                        precision[t,:,k,a,m] = np.array(q)
+        self.eval = {
+            'params': p,
+            'counts': [T, R, K, A, M],
+            'date': datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
+            'precision': precision,
+            'recall':   recall,
+        }
+        toc = time.time()
+        print 'DONE (t=%0.2fs).'%( toc-tic )
+
+    def summarize(self):
+        '''
+        Compute and display summary metrics for evaluation results.
+        Note this functin can *only* be applied on the default parameter setting
+        '''
+        def _summarize( ap=1, iouThr=None, areaRng='all', maxDets=100 ):
+            p = self.params
+            iStr        = ' {:<18} {} @[ IoU={:<9} | area={:>6} | maxDets={:>3} ] = {}'
+            titleStr    = 'Average Precision' if ap == 1 else 'Average Recall'
+            typeStr     = '(AP)' if ap==1 else '(AR)'
+            iouStr      = '%0.2f:%0.2f'%(p.iouThrs[0], p.iouThrs[-1]) if iouThr is None else '%0.2f'%(iouThr)
+            areaStr     = areaRng
+            maxDetsStr  = '%d'%(maxDets)
+
+            aind = [i for i, aRng in enumerate(['all', 'small', 'medium', 'large']) if aRng == areaRng]
+            mind = [i for i, mDet in enumerate([1, 10, 100]) if mDet == maxDets]
+            if ap == 1:
+                # dimension of precision: [TxRxKxAxM]
+                s = self.eval['precision']
+                # IoU
+                if iouThr is not None:
+                    t = np.where(iouThr == p.iouThrs)[0]
+                    s = s[t]
+                # areaRng
+                s = s[:,:,:,aind,mind]
+            else:
+                # dimension of recall: [TxKxAxM]
+                s = self.eval['recall']
+                s = s[:,:,aind,mind]
+            if len(s[s>-1])==0:
+                mean_s = -1
+            else:
+                mean_s = np.mean(s[s>-1])
+            print iStr.format(titleStr, typeStr, iouStr, areaStr, maxDetsStr, '%.3f'%(float(mean_s)))
+            return mean_s
+
+        if not self.eval:
+            raise Exception('Please run accumulate() first')
+        self.stats = np.zeros((12,))
+        self.stats[0] = _summarize(1)
+        self.stats[1] = _summarize(1,iouThr=.5)
+        self.stats[2] = _summarize(1,iouThr=.75)
+        self.stats[3] = _summarize(1,areaRng='small')
+        self.stats[4] = _summarize(1,areaRng='medium')
+        self.stats[5] = _summarize(1,areaRng='large')
+        self.stats[6] = _summarize(0,maxDets=1)
+        self.stats[7] = _summarize(0,maxDets=10)
+        self.stats[8] = _summarize(0,maxDets=100)
+        self.stats[9]  = _summarize(0,areaRng='small')
+        self.stats[10] = _summarize(0,areaRng='medium')
+        self.stats[11] = _summarize(0,areaRng='large')
+
+    def __str__(self):
+        self.summarize()
+
+class Params:
+    '''
+    Params for coco evaluation api
+    '''
+    def __init__(self):
+        self.imgIds = []
+        self.catIds = []
+        # np.arange causes trouble.  the data point on arange is slightly larger than the true value
+        self.iouThrs = np.linspace(.5, 0.95, np.round((0.95-.5)/.05)+1, endpoint=True)
+        self.recThrs = np.linspace(.0, 1.00, np.round((1.00-.0)/.01)+1, endpoint=True)
+        self.maxDets = [1,10,100]
+        self.areaRng = [ [0**2,1e5**2], [0**2, 32**2], [32**2, 96**2], [96**2, 1e5**2] ]
+        self.useSegm = 0
+        self.useCats = 1
\ No newline at end of file
diff --git a/lib/pycocotools/license.txt b/lib/pycocotools/license.txt
new file mode 100644
index 0000000..495c163
--- /dev/null
+++ b/lib/pycocotools/license.txt
@@ -0,0 +1,26 @@
+Copyright (c) 2014, Piotr Dollar and Tsung-Yi Lin
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met: 
+
+1. Redistributions of source code must retain the above copyright notice, this
+   list of conditions and the following disclaimer. 
+2. Redistributions in binary form must reproduce the above copyright notice,
+   this list of conditions and the following disclaimer in the documentation
+   and/or other materials provided with the distribution. 
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
+ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
+ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+The views and conclusions contained in the software and documentation are those
+of the authors and should not be interpreted as representing official policies, 
+either expressed or implied, of the FreeBSD Project.
diff --git a/lib/pycocotools/mask.py b/lib/pycocotools/mask.py
new file mode 100644
index 0000000..6732546
--- /dev/null
+++ b/lib/pycocotools/mask.py
@@ -0,0 +1,82 @@
+__author__ = 'tsungyi'
+
+import pycocotools._mask as _mask
+
+# Interface for manipulating masks stored in RLE format.
+#
+# RLE is a simple yet efficient format for storing binary masks. RLE
+# first divides a vector (or vectorized image) into a series of piecewise
+# constant regions and then for each piece simply stores the length of
+# that piece. For example, given M=[0 0 1 1 1 0 1] the RLE counts would
+# be [2 3 1 1], or for M=[1 1 1 1 1 1 0] the counts would be [0 6 1]
+# (note that the odd counts are always the numbers of zeros). Instead of
+# storing the counts directly, additional compression is achieved with a
+# variable bitrate representation based on a common scheme called LEB128.
+#
+# Compression is greatest given large piecewise constant regions.
+# Specifically, the size of the RLE is proportional to the number of
+# *boundaries* in M (or for an image the number of boundaries in the y
+# direction). Assuming fairly simple shapes, the RLE representation is
+# O(sqrt(n)) where n is number of pixels in the object. Hence space usage
+# is substantially lower, especially for large simple objects (large n).
+#
+# Many common operations on masks can be computed directly using the RLE
+# (without need for decoding). This includes computations such as area,
+# union, intersection, etc. All of these operations are linear in the
+# size of the RLE, in other words they are O(sqrt(n)) where n is the area
+# of the object. Computing these operations on the original mask is O(n).
+# Thus, using the RLE can result in substantial computational savings.
+#
+# The following API functions are defined:
+#  encode         - Encode binary masks using RLE.
+#  decode         - Decode binary masks encoded via RLE.
+#  merge          - Compute union or intersection of encoded masks.
+#  iou            - Compute intersection over union between masks.
+#  area           - Compute area of encoded masks.
+#  toBbox         - Get bounding boxes surrounding encoded masks.
+#  frPyObjects    - Convert polygon, bbox, and uncompressed RLE to encoded RLE mask.
+#
+# Usage:
+#  Rs     = encode( masks )
+#  masks  = decode( Rs )
+#  R      = merge( Rs, intersect=false )
+#  o      = iou( dt, gt, iscrowd )
+#  a      = area( Rs )
+#  bbs    = toBbox( Rs )
+#  Rs     = frPyObjects( [pyObjects], h, w )
+#
+# In the API the following formats are used:
+#  Rs      - [dict] Run-length encoding of binary masks
+#  R       - dict Run-length encoding of binary mask
+#  masks   - [hxwxn] Binary mask(s) (must have type np.ndarray(dtype=uint8) in column-major order)
+#  iscrowd - [nx1] list of np.ndarray. 1 indicates corresponding gt image has crowd region to ignore
+#  bbs     - [nx4] Bounding box(es) stored as [x y w h]
+#  poly    - Polygon stored as [[x1 y1 x2 y2...],[x1 y1 ...],...] (2D list)
+#  dt,gt   - May be either bounding boxes or encoded masks
+# Both poly and bbs are 0-indexed (bbox=[0 0 1 1] encloses first pixel).
+#
+# Finally, a note about the intersection over union (iou) computation.
+# The standard iou of a ground truth (gt) and detected (dt) object is
+#  iou(gt,dt) = area(intersect(gt,dt)) / area(union(gt,dt))
+# For "crowd" regions, we use a modified criteria. If a gt object is
+# marked as "iscrowd", we allow a dt to match any subregion of the gt.
+# Choosing gt' in the crowd gt that best matches the dt can be done using
+# gt'=intersect(dt,gt). Since by definition union(gt',dt)=dt, computing
+#  iou(gt,dt,iscrowd) = iou(gt',dt) = area(intersect(gt,dt)) / area(dt)
+# For crowd gt regions we use this modified criteria above for the iou.
+#
+# To compile run "python setup.py build_ext --inplace"
+# Please do not contact us for help with compiling.
+#
+# Microsoft COCO Toolbox.      version 2.0
+# Data, paper, and tutorials available at:  http://mscoco.org/
+# Code written by Piotr Dollar and Tsung-Yi Lin, 2015.
+# Licensed under the Simplified BSD License [see coco/license.txt]
+
+encode      = _mask.encode
+decode      = _mask.decode
+iou         = _mask.iou
+merge       = _mask.merge
+area        = _mask.area
+toBbox      = _mask.toBbox
+frPyObjects = _mask.frPyObjects
\ No newline at end of file
diff --git a/lib/pycocotools/maskApi.c b/lib/pycocotools/maskApi.c
new file mode 100644
index 0000000..2b2d891
--- /dev/null
+++ b/lib/pycocotools/maskApi.c
@@ -0,0 +1,208 @@
+/**************************************************************************
+* Microsoft COCO Toolbox.      version 2.0
+* Data, paper, and tutorials available at:  http://mscoco.org/
+* Code written by Piotr Dollar and Tsung-Yi Lin, 2015.
+* Licensed under the Simplified BSD License [see coco/license.txt]
+**************************************************************************/
+#include "maskApi.h"
+#include <math.h>
+#include <stdlib.h>
+
+uint umin( uint a, uint b ) { return (a<b) ? a : b; }
+uint umax( uint a, uint b ) { return (a>b) ? a : b; }
+
+void rleInit( RLE *R, siz h, siz w, siz m, uint *cnts ) {
+  R->h=h; R->w=w; R->m=m; R->cnts=(m==0)?0:malloc(sizeof(uint)*m);
+  if(cnts) for(siz j=0; j<m; j++) R->cnts[j]=cnts[j];
+}
+
+void rleFree( RLE *R ) {
+  free(R->cnts); R->cnts=0;
+}
+
+void rlesInit( RLE **R, siz n ) {
+  *R = (RLE*) malloc(sizeof(RLE)*n);
+  for(siz i=0; i<n; i++) rleInit((*R)+i,0,0,0,0);
+}
+
+void rlesFree( RLE **R, siz n ) {
+  for(siz i=0; i<n; i++) rleFree((*R)+i); free(*R); *R=0;
+}
+
+void rleEncode( RLE *R, const byte *M, siz h, siz w, siz n ) {
+  siz i, j, k, a=w*h; uint c, *cnts; byte p;
+  cnts = malloc(sizeof(uint)*(a+1));
+  for(i=0; i<n; i++) {
+    const byte *T=M+a*i; k=0; p=0; c=0;
+    for(j=0; j<a; j++) { if(T[j]!=p) { cnts[k++]=c; c=0; p=T[j]; } c++; }
+    cnts[k++]=c; rleInit(R+i,h,w,k,cnts);
+  }
+  free(cnts);
+}
+
+void rleDecode( const RLE *R, byte *M, siz n ) {
+  for( siz i=0; i<n; i++ ) {
+    byte v=0; for( siz j=0; j<R[i].m; j++ ) {
+      for( siz k=0; k<R[i].cnts[j]; k++ ) *(M++)=v; v=!v; }}
+}
+
+void rleMerge( const RLE *R, RLE *M, siz n, bool intersect ) {
+  uint *cnts, c, ca, cb, cc, ct; bool v, va, vb, vp;
+  siz i, a, b, h=R[0].h, w=R[0].w, m=R[0].m; RLE A, B;
+  if(n==0) { rleInit(M,0,0,0,0); return; }
+  if(n==1) { rleInit(M,h,w,m,R[0].cnts); return; }
+  cnts = malloc(sizeof(uint)*(h*w+1));
+  for( a=0; a<m; a++ ) cnts[a]=R[0].cnts[a];
+  for( i=1; i<n; i++ ) {
+    B=R[i]; if(B.h!=h||B.w!=w) { h=w=m=0; break; }
+    rleInit(&A,h,w,m,cnts); ca=A.cnts[0]; cb=B.cnts[0];
+    v=va=vb=0; m=0; a=b=1; cc=0; ct=1;
+    while( ct>0 ) {
+      c=umin(ca,cb); cc+=c; ct=0;
+      ca-=c; if(!ca && a<A.m) { ca=A.cnts[a++]; va=!va; } ct+=ca;
+      cb-=c; if(!cb && b<B.m) { cb=B.cnts[b++]; vb=!vb; } ct+=cb;
+      vp=v; if(intersect) v=va&&vb; else v=va||vb;
+      if( v!=vp||ct==0 ) { cnts[m++]=cc; cc=0; }
+    }
+    rleFree(&A);
+  }
+  rleInit(M,h,w,m,cnts); free(cnts);
+}
+
+void rleArea( const RLE *R, siz n, uint *a ) {
+  for( siz i=0; i<n; i++ ) {
+    a[i]=0; for( siz j=1; j<R[i].m; j+=2 ) a[i]+=R[i].cnts[j]; }
+}
+
+void rleIou( RLE *dt, RLE *gt, siz m, siz n, byte *iscrowd, double *o ) {
+  siz g, d; BB db, gb; bool crowd;
+  db=malloc(sizeof(double)*m*4); rleToBbox(dt,db,m);
+  gb=malloc(sizeof(double)*n*4); rleToBbox(gt,gb,n);
+  bbIou(db,gb,m,n,iscrowd,o); free(db); free(gb);
+  for( g=0; g<n; g++ ) for( d=0; d<m; d++ ) if(o[g*m+d]>0) {
+    crowd=iscrowd!=NULL && iscrowd[g];
+    if(dt[d].h!=gt[g].h || dt[d].w!=gt[g].w) { o[g*m+d]=-1; continue; }
+    siz ka, kb, a, b; uint c, ca, cb, ct, i, u; bool va, vb;
+    ca=dt[d].cnts[0]; ka=dt[d].m; va=vb=0;
+    cb=gt[g].cnts[0]; kb=gt[g].m; a=b=1; i=u=0; ct=1;
+    while( ct>0 ) {
+      c=umin(ca,cb); if(va||vb) { u+=c; if(va&&vb) i+=c; } ct=0;
+      ca-=c; if(!ca && a<ka) { ca=dt[d].cnts[a++]; va=!va; } ct+=ca;
+      cb-=c; if(!cb && b<kb) { cb=gt[g].cnts[b++]; vb=!vb; } ct+=cb;
+    }
+    if(i==0) u=1; else if(crowd) rleArea(dt+d,1,&u);
+    o[g*m+d] = (double)i/(double)u;
+  }
+}
+
+void bbIou( BB dt, BB gt, siz m, siz n, byte *iscrowd, double *o ) {
+  double h, w, i, u, ga, da; siz g, d; bool crowd;
+  for( g=0; g<n; g++ ) {
+    BB G=gt+g*4; ga=G[2]*G[3]; crowd=iscrowd!=NULL && iscrowd[g];
+    for( d=0; d<m; d++ ) {
+      BB D=dt+d*4; da=D[2]*D[3]; o[g*m+d]=0;
+      w=fmin(D[2]+D[0],G[2]+G[0])-fmax(D[0],G[0]); if(w<=0) continue;
+      h=fmin(D[3]+D[1],G[3]+G[1])-fmax(D[1],G[1]); if(h<=0) continue;
+      i=w*h; u = crowd ? da : da+ga-i; o[g*m+d]=i/u;
+    }
+  }
+}
+
+void rleToBbox( const RLE *R, BB bb, siz n ) {
+  for( siz i=0; i<n; i++ ) {
+    uint h, w, x, y, xs, ys, xe, ye, cc, t; siz j, m;
+    h=(uint)R[i].h; w=(uint)R[i].w; m=R[i].m;
+    m=((siz)(m/2))*2; xs=w; ys=h; xe=ye=0; cc=0;
+    if(m==0) { bb[4*i+0]=bb[4*i+1]=bb[4*i+2]=bb[4*i+3]=0; continue; }
+    for( j=0; j<m; j++ ) {
+      cc+=R[i].cnts[j]; t=cc-j%2; y=t%h; x=(t-y)/h;
+      xs=umin(xs,x); xe=umax(xe,x); ys=umin(ys,y); ye=umax(ye,y);
+    }
+    bb[4*i+0]=xs; bb[4*i+2]=xe-xs+1;
+    bb[4*i+1]=ys; bb[4*i+3]=ye-ys+1;
+  }
+}
+
+void rleFrBbox( RLE *R, const BB bb, siz h, siz w, siz n ) {
+  for( siz i=0; i<n; i++ ) {
+    double xs=bb[4*i+0], xe=xs+bb[4*i+2];
+    double ys=bb[4*i+1], ye=ys+bb[4*i+3];
+    double xy[8] = {xs,ys,xs,ye,xe,ye,xe,ys};
+    rleFrPoly( R+i, xy, 4, h, w );
+  }
+}
+
+int uintCompare(const void *a, const void *b) {
+  uint c=*((uint*)a), d=*((uint*)b); return c>d?1:c<d?-1:0;
+}
+
+void rleFrPoly( RLE *R, const double *xy, siz k, siz h, siz w ) {
+  // upsample and get discrete points densely along entire boundary
+  siz j, m=0; double scale=5; int *x, *y, *u, *v; uint *a, *b;
+  x=malloc(sizeof(int)*(k+1)); y=malloc(sizeof(int)*(k+1));
+  for(j=0; j<k; j++) x[j]=(int)(scale*xy[j*2+0]+.5); x[k]=x[0];
+  for(j=0; j<k; j++) y[j]=(int)(scale*xy[j*2+1]+.5); y[k]=y[0];
+  for(j=0; j<k; j++) m+=umax(abs(x[j]-x[j+1]),abs(y[j]-y[j+1]))+1;
+  u=malloc(sizeof(int)*m); v=malloc(sizeof(int)*m); m=0;
+  for( j=0; j<k; j++ ) {
+    int xs=x[j], xe=x[j+1], ys=y[j], ye=y[j+1], dx, dy, t;
+    bool flip; double s; dx=abs(xe-xs); dy=abs(ys-ye);
+    flip = (dx>=dy && xs>xe) || (dx<dy && ys>ye);
+    if(flip) { t=xs; xs=xe; xe=t; t=ys; ys=ye; ye=t; }
+    s = dx>=dy ? (double)(ye-ys)/dx : (double)(xe-xs)/dy;
+    if(dx>=dy) for( int d=0; d<=dx; d++ ) {
+      t=flip?dx-d:d; u[m]=t+xs; v[m]=(int)(ys+s*t+.5); m++;
+    } else for( int d=0; d<=dy; d++ ) {
+      t=flip?dy-d:d; v[m]=t+ys; u[m]=(int)(xs+s*t+.5); m++;
+    }
+  }
+  // get points along y-boundary and downsample
+  free(x); free(y); k=m; m=0; double xd, yd;
+  x=malloc(sizeof(int)*k); y=malloc(sizeof(int)*k);
+  for( j=1; j<k; j++ ) if(u[j]!=u[j-1]) {
+    xd=(double)(u[j]<u[j-1]?u[j]:u[j]-1); xd=(xd+.5)/scale-.5;
+    if( floor(xd)!=xd || xd<0 || xd>w-1 ) continue;
+    yd=(double)(v[j]<v[j-1]?v[j]:v[j-1]); yd=(yd+.5)/scale-.5;
+    if(yd<0) yd=0; else if(yd>h) yd=h; yd=ceil(yd);
+    x[m]=(int) xd; y[m]=(int) yd; m++;
+  }
+  // compute rle encoding given y-boundary points
+  k=m; a=malloc(sizeof(uint)*(k+1));
+  for( j=0; j<k; j++ ) a[j]=(uint)(x[j]*(int)(h)+y[j]);
+  a[k++]=(uint)(h*w); free(u); free(v); free(x); free(y);
+  qsort(a,k,sizeof(uint),uintCompare); uint p=0;
+  for( j=0; j<k; j++ ) { uint t=a[j]; a[j]-=p; p=t; }
+  b=malloc(sizeof(uint)*k); j=m=0; b[m++]=a[j++];
+  while(j<k) if(a[j]>0) b[m++]=a[j++]; else {
+    j++; if(j<k) b[m-1]+=a[j++]; }
+  rleInit(R,h,w,m,b); free(a); free(b);
+}
+
+char* rleToString( const RLE *R ) {
+  // Similar to LEB128 but using 6 bits/char and ascii chars 48-111.
+  siz i, m=R->m, p=0; long x; bool more;
+  char *s=malloc(sizeof(char)*m*6);
+  for( i=0; i<m; i++ ) {
+    x=(long) R->cnts[i]; if(i>2) x-=(long) R->cnts[i-2]; more=1;
+    while( more ) {
+      char c=x & 0x1f; x >>= 5; more=(c & 0x10) ? x!=-1 : x!=0;
+      if(more) c |= 0x20; c+=48; s[p++]=c;
+    }
+  }
+  s[p]=0; return s;
+}
+
+void rleFrString( RLE *R, char *s, siz h, siz w ) {
+  siz m=0, p=0, k; long x; bool more; uint *cnts;
+  while( s[m] ) m++; cnts=malloc(sizeof(uint)*m); m=0;
+  while( s[p] ) {
+    x=0; k=0; more=1;
+    while( more ) {
+      char c=s[p]-48; x |= (c & 0x1f) << 5*k;
+      more = c & 0x20; p++; k++;
+      if(!more && (c & 0x10)) x |= -1 << 5*k;
+    }
+    if(m>2) x+=(long) cnts[m-2]; cnts[m++]=(uint) x;
+  }
+  rleInit(R,h,w,m,cnts); free(cnts);
+}
diff --git a/lib/pycocotools/maskApi.h b/lib/pycocotools/maskApi.h
new file mode 100644
index 0000000..ff16116
--- /dev/null
+++ b/lib/pycocotools/maskApi.h
@@ -0,0 +1,55 @@
+/**************************************************************************
+* Microsoft COCO Toolbox.      version 2.0
+* Data, paper, and tutorials available at:  http://mscoco.org/
+* Code written by Piotr Dollar and Tsung-Yi Lin, 2015.
+* Licensed under the Simplified BSD License [see coco/license.txt]
+**************************************************************************/
+#pragma once
+#include <stdbool.h>
+
+typedef unsigned int uint;
+typedef unsigned long siz;
+typedef unsigned char byte;
+typedef double* BB;
+typedef struct { siz h, w, m; uint *cnts; } RLE;
+
+// Initialize/destroy RLE.
+void rleInit( RLE *R, siz h, siz w, siz m, uint *cnts );
+void rleFree( RLE *R );
+
+// Initialize/destroy RLE array.
+void rlesInit( RLE **R, siz n );
+void rlesFree( RLE **R, siz n );
+
+// Encode binary masks using RLE.
+void rleEncode( RLE *R, const byte *mask, siz h, siz w, siz n );
+
+// Decode binary masks encoded via RLE.
+void rleDecode( const RLE *R, byte *mask, siz n );
+
+// Compute union or intersection of encoded masks.
+void rleMerge( const RLE *R, RLE *M, siz n, bool intersect );
+
+// Compute area of encoded masks.
+void rleArea( const RLE *R, siz n, uint *a );
+
+// Compute intersection over union between masks.
+void rleIou( RLE *dt, RLE *gt, siz m, siz n, byte *iscrowd, double *o );
+
+// Compute intersection over union between bounding boxes.
+void bbIou( BB dt, BB gt, siz m, siz n, byte *iscrowd, double *o );
+
+// Get bounding boxes surrounding encoded masks.
+void rleToBbox( const RLE *R, BB bb, siz n );
+
+// Convert bounding boxes to encoded masks.
+void rleFrBbox( RLE *R, const BB bb, siz h, siz w, siz n );
+
+// Convert polygon to encoded mask.
+void rleFrPoly( RLE *R, const double *xy, siz k, siz h, siz w );
+
+// Get compressed string representation of encoded mask.
+char* rleToString( const RLE *R );
+
+// Convert from compressed string representation of encoded mask.
+void rleFrString( RLE *R, char *s, siz h, siz w );
diff --git a/lib/roi_data_layer/__init__.py b/lib/roi_data_layer/__init__.py
new file mode 100644
index 0000000..7ba6a65
--- /dev/null
+++ b/lib/roi_data_layer/__init__.py
@@ -0,0 +1,6 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
diff --git a/lib/roi_data_layer/layer.py b/lib/roi_data_layer/layer.py
new file mode 100644
index 0000000..04f4172
--- /dev/null
+++ b/lib/roi_data_layer/layer.py
@@ -0,0 +1,196 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""The data layer used during training to train a Fast R-CNN network.
+
+RoIDataLayer implements a Caffe Python layer.
+"""
+
+import caffe
+from fast_rcnn.config import cfg
+from roi_data_layer.minibatch import get_minibatch
+import numpy as np
+import yaml
+from multiprocessing import Process, Queue
+
+class RoIDataLayer(caffe.Layer):
+    """Fast R-CNN data layer used for training."""
+
+    def _shuffle_roidb_inds(self):
+        """Randomly permute the training roidb."""
+        if cfg.TRAIN.ASPECT_GROUPING:
+            widths = np.array([r['width'] for r in self._roidb])
+            heights = np.array([r['height'] for r in self._roidb])
+            horz = (widths >= heights)
+            vert = np.logical_not(horz)
+            horz_inds = np.where(horz)[0]
+            vert_inds = np.where(vert)[0]
+            inds = np.hstack((
+                np.random.permutation(horz_inds),
+                np.random.permutation(vert_inds)))
+            inds = np.reshape(inds, (-1, 2))
+            row_perm = np.random.permutation(np.arange(inds.shape[0]))
+            inds = np.reshape(inds[row_perm, :], (-1,))
+            self._perm = inds
+        else:
+            self._perm = np.random.permutation(np.arange(len(self._roidb)))
+        self._cur = 0
+
+    def _get_next_minibatch_inds(self):
+        """Return the roidb indices for the next minibatch."""
+        if self._cur + cfg.TRAIN.IMS_PER_BATCH >= len(self._roidb):
+            self._shuffle_roidb_inds()
+
+        db_inds = self._perm[self._cur:self._cur + cfg.TRAIN.IMS_PER_BATCH]
+        self._cur += cfg.TRAIN.IMS_PER_BATCH
+        return db_inds
+
+    def _get_next_minibatch(self):
+        """Return the blobs to be used for the next minibatch.
+
+        If cfg.TRAIN.USE_PREFETCH is True, then blobs will be computed in a
+        separate process and made available through self._blob_queue.
+        """
+        if cfg.TRAIN.USE_PREFETCH:
+            return self._blob_queue.get()
+        else:
+            db_inds = self._get_next_minibatch_inds()
+            minibatch_db = [self._roidb[i] for i in db_inds]
+            return get_minibatch(minibatch_db, self._num_classes)
+
+    def set_roidb(self, roidb):
+        """Set the roidb to be used by this layer during training."""
+        self._roidb = roidb
+        self._shuffle_roidb_inds()
+        if cfg.TRAIN.USE_PREFETCH:
+            self._blob_queue = Queue(10)
+            self._prefetch_process = BlobFetcher(self._blob_queue,
+                                                 self._roidb,
+                                                 self._num_classes)
+            self._prefetch_process.start()
+            # Terminate the child process when the parent exists
+            def cleanup():
+                print 'Terminating BlobFetcher'
+                self._prefetch_process.terminate()
+                self._prefetch_process.join()
+            import atexit
+            atexit.register(cleanup)
+
+    def setup(self, bottom, top):
+        """Setup the RoIDataLayer."""
+
+        # parse the layer parameter string, which must be valid YAML
+        layer_params = yaml.load(self.param_str)
+
+        self._num_classes = layer_params['num_classes']
+
+        self._name_to_top_map = {}
+
+        # data blob: holds a batch of N images, each with 3 channels
+        idx = 0
+        top[idx].reshape(cfg.TRAIN.IMS_PER_BATCH, 3,
+            max(cfg.TRAIN.SCALES), cfg.TRAIN.MAX_SIZE)
+        self._name_to_top_map['data'] = idx
+        idx += 1
+
+        if cfg.TRAIN.HAS_RPN:
+            top[idx].reshape(1, 3)
+            self._name_to_top_map['im_info'] = idx
+            idx += 1
+
+            top[idx].reshape(1, 4)
+            self._name_to_top_map['gt_boxes'] = idx
+            idx += 1
+        else: # not using RPN
+            # rois blob: holds R regions of interest, each is a 5-tuple
+            # (n, x1, y1, x2, y2) specifying an image batch index n and a
+            # rectangle (x1, y1, x2, y2)
+            top[idx].reshape(1, 5)
+            self._name_to_top_map['rois'] = idx
+            idx += 1
+
+            # labels blob: R categorical labels in [0, ..., K] for K foreground
+            # classes plus background
+            top[idx].reshape(1)
+            self._name_to_top_map['labels'] = idx
+            idx += 1
+
+            if cfg.TRAIN.BBOX_REG:
+                # bbox_targets blob: R bounding-box regression targets with 4
+                # targets per class
+                top[idx].reshape(1, self._num_classes * 4)
+                self._name_to_top_map['bbox_targets'] = idx
+                idx += 1
+
+                # bbox_inside_weights blob: At most 4 targets per roi are active;
+                # thisbinary vector sepcifies the subset of active targets
+                top[idx].reshape(1, self._num_classes * 4)
+                self._name_to_top_map['bbox_inside_weights'] = idx
+                idx += 1
+
+                top[idx].reshape(1, self._num_classes * 4)
+                self._name_to_top_map['bbox_outside_weights'] = idx
+                idx += 1
+
+        print 'RoiDataLayer: name_to_top:', self._name_to_top_map
+        assert len(top) == len(self._name_to_top_map)
+
+    def forward(self, bottom, top):
+        """Get blobs and copy them into this layer's top blob vector."""
+        blobs = self._get_next_minibatch()
+
+        for blob_name, blob in blobs.iteritems():
+            top_ind = self._name_to_top_map[blob_name]
+            # Reshape net's input blobs
+            top[top_ind].reshape(*(blob.shape))
+            # Copy data into net's input blobs
+            top[top_ind].data[...] = blob.astype(np.float32, copy=False)
+
+    def backward(self, top, propagate_down, bottom):
+        """This layer does not propagate gradients."""
+        pass
+
+    def reshape(self, bottom, top):
+        """Reshaping happens during the call to forward."""
+        pass
+
+class BlobFetcher(Process):
+    """Experimental class for prefetching blobs in a separate process."""
+    def __init__(self, queue, roidb, num_classes):
+        super(BlobFetcher, self).__init__()
+        self._queue = queue
+        self._roidb = roidb
+        self._num_classes = num_classes
+        self._perm = None
+        self._cur = 0
+        self._shuffle_roidb_inds()
+        # fix the random seed for reproducibility
+        np.random.seed(cfg.RNG_SEED)
+
+    def _shuffle_roidb_inds(self):
+        """Randomly permute the training roidb."""
+        # TODO(rbg): remove duplicated code
+        self._perm = np.random.permutation(np.arange(len(self._roidb)))
+        self._cur = 0
+
+    def _get_next_minibatch_inds(self):
+        """Return the roidb indices for the next minibatch."""
+        # TODO(rbg): remove duplicated code
+        if self._cur + cfg.TRAIN.IMS_PER_BATCH >= len(self._roidb):
+            self._shuffle_roidb_inds()
+
+        db_inds = self._perm[self._cur:self._cur + cfg.TRAIN.IMS_PER_BATCH]
+        self._cur += cfg.TRAIN.IMS_PER_BATCH
+        return db_inds
+
+    def run(self):
+        print 'BlobFetcher started'
+        while True:
+            db_inds = self._get_next_minibatch_inds()
+            minibatch_db = [self._roidb[i] for i in db_inds]
+            blobs = get_minibatch(minibatch_db, self._num_classes)
+            self._queue.put(blobs)
diff --git a/lib/roi_data_layer/minibatch.py b/lib/roi_data_layer/minibatch.py
new file mode 100644
index 0000000..f4535b0
--- /dev/null
+++ b/lib/roi_data_layer/minibatch.py
@@ -0,0 +1,199 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Compute minibatch blobs for training a Fast R-CNN network."""
+
+import numpy as np
+import numpy.random as npr
+import cv2
+from fast_rcnn.config import cfg
+from utils.blob import prep_im_for_blob, im_list_to_blob
+
+def get_minibatch(roidb, num_classes):
+    """Given a roidb, construct a minibatch sampled from it."""
+    num_images = len(roidb)
+    # Sample random scales to use for each image in this batch
+    random_scale_inds = npr.randint(0, high=len(cfg.TRAIN.SCALES),
+                                    size=num_images)
+    assert(cfg.TRAIN.BATCH_SIZE % num_images == 0), \
+        'num_images ({}) must divide BATCH_SIZE ({})'. \
+        format(num_images, cfg.TRAIN.BATCH_SIZE)
+    rois_per_image = cfg.TRAIN.BATCH_SIZE / num_images
+    fg_rois_per_image = np.round(cfg.TRAIN.FG_FRACTION * rois_per_image)
+
+    # Get the input image blob, formatted for caffe
+    im_blob, im_scales = _get_image_blob(roidb, random_scale_inds)
+
+    blobs = {'data': im_blob}
+
+    if cfg.TRAIN.HAS_RPN:
+        assert len(im_scales) == 1, "Single batch only"
+        assert len(roidb) == 1, "Single batch only"
+        # gt boxes: (x1, y1, x2, y2, cls)
+        gt_inds = np.where(roidb[0]['gt_classes'] != 0)[0]
+        gt_boxes = np.empty((len(gt_inds), 5), dtype=np.float32)
+        gt_boxes[:, 0:4] = roidb[0]['boxes'][gt_inds, :] * im_scales[0]
+        gt_boxes[:, 4] = roidb[0]['gt_classes'][gt_inds]
+        blobs['gt_boxes'] = gt_boxes
+        blobs['im_info'] = np.array(
+            [[im_blob.shape[2], im_blob.shape[3], im_scales[0]]],
+            dtype=np.float32)
+    else: # not using RPN
+        # Now, build the region of interest and label blobs
+        rois_blob = np.zeros((0, 5), dtype=np.float32)
+        labels_blob = np.zeros((0), dtype=np.float32)
+        bbox_targets_blob = np.zeros((0, 4 * num_classes), dtype=np.float32)
+        bbox_inside_blob = np.zeros(bbox_targets_blob.shape, dtype=np.float32)
+        # all_overlaps = []
+        for im_i in xrange(num_images):
+            labels, overlaps, im_rois, bbox_targets, bbox_inside_weights \
+                = _sample_rois(roidb[im_i], fg_rois_per_image, rois_per_image,
+                               num_classes)
+
+            # Add to RoIs blob
+            rois = _project_im_rois(im_rois, im_scales[im_i])
+            batch_ind = im_i * np.ones((rois.shape[0], 1))
+            rois_blob_this_image = np.hstack((batch_ind, rois))
+            rois_blob = np.vstack((rois_blob, rois_blob_this_image))
+
+            # Add to labels, bbox targets, and bbox loss blobs
+            labels_blob = np.hstack((labels_blob, labels))
+            bbox_targets_blob = np.vstack((bbox_targets_blob, bbox_targets))
+            bbox_inside_blob = np.vstack((bbox_inside_blob, bbox_inside_weights))
+            # all_overlaps = np.hstack((all_overlaps, overlaps))
+
+        # For debug visualizations
+        # _vis_minibatch(im_blob, rois_blob, labels_blob, all_overlaps)
+
+        blobs['rois'] = rois_blob
+        blobs['labels'] = labels_blob
+
+        if cfg.TRAIN.BBOX_REG:
+            blobs['bbox_targets'] = bbox_targets_blob
+            blobs['bbox_inside_weights'] = bbox_inside_blob
+            blobs['bbox_outside_weights'] = \
+                np.array(bbox_inside_blob > 0).astype(np.float32)
+
+    return blobs
+
+def _sample_rois(roidb, fg_rois_per_image, rois_per_image, num_classes):
+    """Generate a random sample of RoIs comprising foreground and background
+    examples.
+    """
+    # label = class RoI has max overlap with
+    labels = roidb['max_classes']
+    overlaps = roidb['max_overlaps']
+    rois = roidb['boxes']
+
+    # Select foreground RoIs as those with >= FG_THRESH overlap
+    fg_inds = np.where(overlaps >= cfg.TRAIN.FG_THRESH)[0]
+    # Guard against the case when an image has fewer than fg_rois_per_image
+    # foreground RoIs
+    fg_rois_per_this_image = np.minimum(fg_rois_per_image, fg_inds.size)
+    # Sample foreground regions without replacement
+    if fg_inds.size > 0:
+        fg_inds = npr.choice(
+                fg_inds, size=fg_rois_per_this_image, replace=False)
+
+    # Select background RoIs as those within [BG_THRESH_LO, BG_THRESH_HI)
+    bg_inds = np.where((overlaps < cfg.TRAIN.BG_THRESH_HI) &
+                       (overlaps >= cfg.TRAIN.BG_THRESH_LO))[0]
+    # Compute number of background RoIs to take from this image (guarding
+    # against there being fewer than desired)
+    bg_rois_per_this_image = rois_per_image - fg_rois_per_this_image
+    bg_rois_per_this_image = np.minimum(bg_rois_per_this_image,
+                                        bg_inds.size)
+    # Sample foreground regions without replacement
+    if bg_inds.size > 0:
+        bg_inds = npr.choice(
+                bg_inds, size=bg_rois_per_this_image, replace=False)
+
+    # The indices that we're selecting (both fg and bg)
+    keep_inds = np.append(fg_inds, bg_inds)
+    # Select sampled values from various arrays:
+    labels = labels[keep_inds]
+    # Clamp labels for the background RoIs to 0
+    labels[fg_rois_per_this_image:] = 0
+    overlaps = overlaps[keep_inds]
+    rois = rois[keep_inds]
+
+    bbox_targets, bbox_inside_weights = _get_bbox_regression_labels(
+            roidb['bbox_targets'][keep_inds, :], num_classes)
+
+    return labels, overlaps, rois, bbox_targets, bbox_inside_weights
+
+def _get_image_blob(roidb, scale_inds):
+    """Builds an input blob from the images in the roidb at the specified
+    scales.
+    """
+    num_images = len(roidb)
+    processed_ims = []
+    im_scales = []
+    for i in xrange(num_images):
+        im = cv2.imread(roidb[i]['image'])
+        if roidb[i]['flipped']:
+            im = im[:, ::-1, :]
+        target_size = cfg.TRAIN.SCALES[scale_inds[i]]
+        im, im_scale = prep_im_for_blob(im, cfg.PIXEL_MEANS, target_size,
+                                        cfg.TRAIN.MAX_SIZE)
+        im_scales.append(im_scale)
+        processed_ims.append(im)
+
+    # Create a blob to hold the input images
+    blob = im_list_to_blob(processed_ims)
+
+    return blob, im_scales
+
+def _project_im_rois(im_rois, im_scale_factor):
+    """Project image RoIs into the rescaled training image."""
+    rois = im_rois * im_scale_factor
+    return rois
+
+def _get_bbox_regression_labels(bbox_target_data, num_classes):
+    """Bounding-box regression targets are stored in a compact form in the
+    roidb.
+
+    This function expands those targets into the 4-of-4*K representation used
+    by the network (i.e. only one class has non-zero targets). The loss weights
+    are similarly expanded.
+
+    Returns:
+        bbox_target_data (ndarray): N x 4K blob of regression targets
+        bbox_inside_weights (ndarray): N x 4K blob of loss weights
+    """
+    clss = bbox_target_data[:, 0]
+    bbox_targets = np.zeros((clss.size, 4 * num_classes), dtype=np.float32)
+    bbox_inside_weights = np.zeros(bbox_targets.shape, dtype=np.float32)
+    inds = np.where(clss > 0)[0]
+    for ind in inds:
+        cls = clss[ind]
+        start = 4 * cls
+        end = start + 4
+        bbox_targets[ind, start:end] = bbox_target_data[ind, 1:]
+        bbox_inside_weights[ind, start:end] = cfg.TRAIN.BBOX_INSIDE_WEIGHTS
+    return bbox_targets, bbox_inside_weights
+
+def _vis_minibatch(im_blob, rois_blob, labels_blob, overlaps):
+    """Visualize a mini-batch for debugging."""
+    import matplotlib.pyplot as plt
+    for i in xrange(rois_blob.shape[0]):
+        rois = rois_blob[i, :]
+        im_ind = rois[0]
+        roi = rois[1:]
+        im = im_blob[im_ind, :, :, :].transpose((1, 2, 0)).copy()
+        im += cfg.PIXEL_MEANS
+        im = im[:, :, (2, 1, 0)]
+        im = im.astype(np.uint8)
+        cls = labels_blob[i]
+        plt.imshow(im)
+        print 'class: ', cls, ' overlap: ', overlaps[i]
+        plt.gca().add_patch(
+            plt.Rectangle((roi[0], roi[1]), roi[2] - roi[0],
+                          roi[3] - roi[1], fill=False,
+                          edgecolor='r', linewidth=3)
+            )
+        plt.show()
diff --git a/lib/roi_data_layer/roidb.py b/lib/roi_data_layer/roidb.py
new file mode 100644
index 0000000..f2bd231
--- /dev/null
+++ b/lib/roi_data_layer/roidb.py
@@ -0,0 +1,133 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Transform a roidb into a trainable roidb by adding a bunch of metadata."""
+
+import numpy as np
+from fast_rcnn.config import cfg
+from fast_rcnn.bbox_transform import bbox_transform
+from utils.cython_bbox import bbox_overlaps
+import PIL
+
+def prepare_roidb(imdb):
+    """Enrich the imdb's roidb by adding some derived quantities that
+    are useful for training. This function precomputes the maximum
+    overlap, taken over ground-truth boxes, between each ROI and
+    each ground-truth box. The class with maximum overlap is also
+    recorded.
+    """
+    sizes = [PIL.Image.open(imdb.image_path_at(i)).size
+             for i in xrange(imdb.num_images)]
+    roidb = imdb.roidb
+    for i in xrange(len(imdb.image_index)):
+        roidb[i]['image'] = imdb.image_path_at(i)
+        roidb[i]['width'] = sizes[i][0]
+        roidb[i]['height'] = sizes[i][1]
+        # need gt_overlaps as a dense array for argmax
+        gt_overlaps = roidb[i]['gt_overlaps'].toarray()
+        # max overlap with gt over classes (columns)
+        max_overlaps = gt_overlaps.max(axis=1)
+        # gt class that had the max overlap
+        max_classes = gt_overlaps.argmax(axis=1)
+        roidb[i]['max_classes'] = max_classes
+        roidb[i]['max_overlaps'] = max_overlaps
+        # sanity checks
+        # max overlap of 0 => class should be zero (background)
+        zero_inds = np.where(max_overlaps == 0)[0]
+        assert all(max_classes[zero_inds] == 0)
+        # max overlap > 0 => class should not be zero (must be a fg class)
+        nonzero_inds = np.where(max_overlaps > 0)[0]
+        assert all(max_classes[nonzero_inds] != 0)
+
+def add_bbox_regression_targets(roidb):
+    """Add information needed to train bounding-box regressors."""
+    assert len(roidb) > 0
+    assert 'max_classes' in roidb[0], 'Did you call prepare_roidb first?'
+
+    num_images = len(roidb)
+    # Infer number of classes from the number of columns in gt_overlaps
+    num_classes = 2 if cfg.TRAIN.AGONISTIC else roidb[0]['gt_overlaps'].shape[1]
+    for im_i in xrange(num_images):
+        rois = roidb[im_i]['boxes']
+        max_overlaps = roidb[im_i]['max_overlaps']
+        max_classes = roidb[im_i]['max_classes']
+        roidb[im_i]['bbox_targets'] = \
+                _compute_targets(rois, max_overlaps, max_classes)
+
+    if cfg.TRAIN.BBOX_NORMALIZE_TARGETS_PRECOMPUTED:
+        # Use fixed / precomputed "means" and "stds" instead of empirical values
+        means = np.tile(
+                np.array(cfg.TRAIN.BBOX_NORMALIZE_MEANS), (num_classes, 1))
+        stds = np.tile(
+                np.array(cfg.TRAIN.BBOX_NORMALIZE_STDS), (num_classes, 1))
+    else:
+        # Compute values needed for means and stds
+        # var(x) = E(x^2) - E(x)^2
+        class_counts = np.zeros((num_classes, 1)) + cfg.EPS
+        sums = np.zeros((num_classes, 4))
+        squared_sums = np.zeros((num_classes, 4))
+        for im_i in xrange(num_images):
+            targets = roidb[im_i]['bbox_targets']
+            for cls in xrange(1, num_classes):
+                cls_inds = np.where(targets[:, 0] == cls)[0]
+                if cls_inds.size > 0:
+                    class_counts[cls] += cls_inds.size
+                    sums[cls, :] += targets[cls_inds, 1:].sum(axis=0)
+                    squared_sums[cls, :] += \
+                            (targets[cls_inds, 1:] ** 2).sum(axis=0)
+
+        means = sums / class_counts
+        stds = np.sqrt(squared_sums / class_counts - means ** 2)
+
+    print 'bbox target means:'
+    print means
+    print means[1:, :].mean(axis=0) # ignore bg class
+    print 'bbox target stdevs:'
+    print stds
+    print stds[1:, :].mean(axis=0) # ignore bg class
+
+    # Normalize targets
+    if cfg.TRAIN.BBOX_NORMALIZE_TARGETS:
+        print "Normalizing targets"
+        for im_i in xrange(num_images):
+            targets = roidb[im_i]['bbox_targets']
+            for cls in xrange(1, num_classes):
+                cls_inds = np.where(targets[:, 0] == cls)[0]
+                roidb[im_i]['bbox_targets'][cls_inds, 1:] -= means[cls, :]
+                roidb[im_i]['bbox_targets'][cls_inds, 1:] /= stds[cls, :]
+    else:
+        print "NOT normalizing targets"
+
+    # These values will be needed for making predictions
+    # (the predicts will need to be unnormalized and uncentered)
+    return means.ravel(), stds.ravel()
+
+def _compute_targets(rois, overlaps, labels):
+    """Compute bounding-box regression targets for an image."""
+    # Indices of ground-truth ROIs
+    gt_inds = np.where(overlaps == 1)[0]
+    if len(gt_inds) == 0:
+        # Bail if the image has no ground-truth ROIs
+        return np.zeros((rois.shape[0], 5), dtype=np.float32)
+    # Indices of examples for which we try to make predictions
+    ex_inds = np.where(overlaps >= cfg.TRAIN.BBOX_THRESH)[0]
+
+    # Get IoU overlap between each ex ROI and gt ROI
+    ex_gt_overlaps = bbox_overlaps(
+        np.ascontiguousarray(rois[ex_inds, :], dtype=np.float),
+        np.ascontiguousarray(rois[gt_inds, :], dtype=np.float))
+
+    # Find which gt ROI each ex ROI has max overlap with:
+    # this will be the ex ROI's gt target
+    gt_assignment = ex_gt_overlaps.argmax(axis=1)
+    gt_rois = rois[gt_inds[gt_assignment], :]
+    ex_rois = rois[ex_inds, :]
+
+    targets = np.zeros((rois.shape[0], 5), dtype=np.float32)
+    targets[ex_inds, 0] = labels[ex_inds]
+    targets[ex_inds, 1:] = bbox_transform(ex_rois, gt_rois)
+    return targets
diff --git a/lib/rpn/README.md b/lib/rpn/README.md
new file mode 100644
index 0000000..80abf48
--- /dev/null
+++ b/lib/rpn/README.md
@@ -0,0 +1,23 @@
+### `rpn` module overview
+
+##### `generate_anchors.py`
+
+Generates a regular grid of multi-scale, multi-aspect anchor boxes.
+
+##### `proposal_layer.py`
+
+Converts RPN outputs (per-anchor scores and bbox regression estimates) into object proposals.
+
+##### `anchor_target_layer.py` 
+
+Generates training targets/labels for each anchor. Classification labels are 1 (object), 0 (not object) or -1 (ignore).
+Bbox regression targets are specified when the classification label is > 0.
+
+##### `proposal_target_layer.py`
+
+Generates training targets/labels for each object proposal: classification labels 0 - K (bg or object class 1, ... , K)
+and bbox regression targets in that case that the label is > 0.
+
+##### `generate.py`
+
+Generate object detection proposals from an imdb using an RPN.
diff --git a/lib/rpn/__init__.py b/lib/rpn/__init__.py
new file mode 100644
index 0000000..23b31b2
--- /dev/null
+++ b/lib/rpn/__init__.py
@@ -0,0 +1,6 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick and Sean Bell
+# --------------------------------------------------------
diff --git a/lib/rpn/anchor_target_layer.py b/lib/rpn/anchor_target_layer.py
new file mode 100644
index 0000000..3934cdd
--- /dev/null
+++ b/lib/rpn/anchor_target_layer.py
@@ -0,0 +1,281 @@
+# --------------------------------------------------------
+# Faster R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick and Sean Bell
+# --------------------------------------------------------
+
+import os
+import caffe
+import yaml
+from fast_rcnn.config import cfg
+import numpy as np
+import numpy.random as npr
+from generate_anchors import generate_anchors
+from utils.cython_bbox import bbox_overlaps
+from fast_rcnn.bbox_transform import bbox_transform
+
+DEBUG = False
+
+class AnchorTargetLayer(caffe.Layer):
+    """
+    Assign anchors to ground-truth targets. Produces anchor classification
+    labels and bounding-box regression targets.
+    """
+
+    def setup(self, bottom, top):
+        layer_params = yaml.load(self.param_str)
+        anchor_scales = layer_params.get('scales', (8, 16, 32))
+        self._anchors = generate_anchors(scales=np.array(anchor_scales))
+        self._num_anchors = self._anchors.shape[0]
+        self._feat_stride = layer_params['feat_stride']
+
+        if DEBUG:
+            print 'anchors:'
+            print self._anchors
+            print 'anchor shapes:'
+            print np.hstack((
+                self._anchors[:, 2::4] - self._anchors[:, 0::4],
+                self._anchors[:, 3::4] - self._anchors[:, 1::4],
+            ))
+            self._counts = cfg.EPS
+            self._sums = np.zeros((1, 4))
+            self._squared_sums = np.zeros((1, 4))
+            self._fg_sum = 0
+            self._bg_sum = 0
+            self._count = 0
+
+        # allow boxes to sit over the edge by a small amount
+        self._allowed_border = layer_params.get('allowed_border', 0)
+
+        height, width = bottom[0].data.shape[-2:]
+        if DEBUG:
+            print 'AnchorTargetLayer: height', height, 'width', width
+
+        A = self._num_anchors
+        # labels
+        top[0].reshape(1, 1, A * height, width)
+        # bbox_targets
+        top[1].reshape(1, A * 4, height, width)
+        # bbox_inside_weights
+        top[2].reshape(1, A * 4, height, width)
+        # bbox_outside_weights
+        top[3].reshape(1, A * 4, height, width)
+
+    def forward(self, bottom, top):
+        # Algorithm:
+        #
+        # for each (H, W) location i
+        #   generate 9 anchor boxes centered on cell i
+        #   apply predicted bbox deltas at cell i to each of the 9 anchors
+        # filter out-of-image anchors
+        # measure GT overlap
+
+        assert bottom[0].data.shape[0] == 1, \
+            'Only single item batches are supported'
+
+        # map of shape (..., H, W)
+        height, width = bottom[0].data.shape[-2:]
+        # GT boxes (x1, y1, x2, y2, label)
+        gt_boxes = bottom[1].data
+        # im_info
+        im_info = bottom[2].data[0, :]
+
+        if DEBUG:
+            print ''
+            print 'im_size: ({}, {})'.format(im_info[0], im_info[1])
+            print 'scale: {}'.format(im_info[2])
+            print 'height, width: ({}, {})'.format(height, width)
+            print 'rpn: gt_boxes.shape', gt_boxes.shape
+            print 'rpn: gt_boxes', gt_boxes
+
+        # 1. Generate proposals from bbox deltas and shifted anchors
+        shift_x = np.arange(0, width) * self._feat_stride
+        shift_y = np.arange(0, height) * self._feat_stride
+        shift_x, shift_y = np.meshgrid(shift_x, shift_y)
+        shifts = np.vstack((shift_x.ravel(), shift_y.ravel(),
+                            shift_x.ravel(), shift_y.ravel())).transpose()
+        # add A anchors (1, A, 4) to
+        # cell K shifts (K, 1, 4) to get
+        # shift anchors (K, A, 4)
+        # reshape to (K*A, 4) shifted anchors
+        A = self._num_anchors
+        K = shifts.shape[0]
+        all_anchors = (self._anchors.reshape((1, A, 4)) +
+                       shifts.reshape((1, K, 4)).transpose((1, 0, 2)))
+        all_anchors = all_anchors.reshape((K * A, 4))
+        total_anchors = int(K * A)
+
+        # only keep anchors inside the image
+        inds_inside = np.where(
+            (all_anchors[:, 0] >= -self._allowed_border) &
+            (all_anchors[:, 1] >= -self._allowed_border) &
+            (all_anchors[:, 2] < im_info[1] + self._allowed_border) &  # width
+            (all_anchors[:, 3] < im_info[0] + self._allowed_border)    # height
+        )[0]
+
+        if DEBUG:
+            print 'total_anchors', total_anchors
+            print 'inds_inside', len(inds_inside)
+
+        # keep only inside anchors
+        anchors = all_anchors[inds_inside, :]
+        if DEBUG:
+            print 'anchors.shape', anchors.shape
+
+        # label: 1 is positive, 0 is negative, -1 is dont care
+        labels = np.empty((len(inds_inside), ), dtype=np.float32)
+        labels.fill(-1)
+
+        # overlaps between the anchors and the gt boxes
+        # overlaps (ex, gt)
+        overlaps = bbox_overlaps(
+            np.ascontiguousarray(anchors, dtype=np.float),
+            np.ascontiguousarray(gt_boxes, dtype=np.float))
+        argmax_overlaps = overlaps.argmax(axis=1)
+        max_overlaps = overlaps[np.arange(len(inds_inside)), argmax_overlaps]
+        gt_argmax_overlaps = overlaps.argmax(axis=0)
+        gt_max_overlaps = overlaps[gt_argmax_overlaps,
+                                   np.arange(overlaps.shape[1])]
+        gt_argmax_overlaps = np.where(overlaps == gt_max_overlaps)[0]
+
+        if not cfg.TRAIN.RPN_CLOBBER_POSITIVES:
+            # assign bg labels first so that positive labels can clobber them
+            labels[max_overlaps < cfg.TRAIN.RPN_NEGATIVE_OVERLAP] = 0
+
+        # fg label: for each gt, anchor with highest overlap
+        labels[gt_argmax_overlaps] = 1
+
+        # fg label: above threshold IOU
+        labels[max_overlaps >= cfg.TRAIN.RPN_POSITIVE_OVERLAP] = 1
+
+        if cfg.TRAIN.RPN_CLOBBER_POSITIVES:
+            # assign bg labels last so that negative labels can clobber positives
+            labels[max_overlaps < cfg.TRAIN.RPN_NEGATIVE_OVERLAP] = 0
+
+        # subsample positive labels if we have too many
+        num_fg = int(cfg.TRAIN.RPN_FG_FRACTION * cfg.TRAIN.RPN_BATCHSIZE)
+        fg_inds = np.where(labels == 1)[0]
+        if len(fg_inds) > num_fg:
+            disable_inds = npr.choice(
+                fg_inds, size=(len(fg_inds) - num_fg), replace=False)
+            labels[disable_inds] = -1
+
+        # subsample negative labels if we have too many
+        num_bg = cfg.TRAIN.RPN_BATCHSIZE - np.sum(labels == 1)
+        bg_inds = np.where(labels == 0)[0]
+        if len(bg_inds) > num_bg:
+            disable_inds = npr.choice(
+                bg_inds, size=(len(bg_inds) - num_bg), replace=False)
+            labels[disable_inds] = -1
+            #print "was %s inds, disabling %s, now %s inds" % (
+                #len(bg_inds), len(disable_inds), np.sum(labels == 0))
+
+        bbox_targets = np.zeros((len(inds_inside), 4), dtype=np.float32)
+        bbox_targets = _compute_targets(anchors, gt_boxes[argmax_overlaps, :])
+
+        bbox_inside_weights = np.zeros((len(inds_inside), 4), dtype=np.float32)
+        bbox_inside_weights[labels == 1, :] = np.array(cfg.TRAIN.RPN_BBOX_INSIDE_WEIGHTS)
+
+        bbox_outside_weights = np.zeros((len(inds_inside), 4), dtype=np.float32)
+        if cfg.TRAIN.RPN_POSITIVE_WEIGHT < 0:
+            # uniform weighting of examples (given non-uniform sampling)
+            num_examples = np.sum(labels >= 0)
+            positive_weights = np.ones((1, 4)) * 1.0 / num_examples
+            negative_weights = np.ones((1, 4)) * 1.0 / num_examples
+        else:
+            assert ((cfg.TRAIN.RPN_POSITIVE_WEIGHT > 0) &
+                    (cfg.TRAIN.RPN_POSITIVE_WEIGHT < 1))
+            positive_weights = (cfg.TRAIN.RPN_POSITIVE_WEIGHT /
+                                np.sum(labels == 1))
+            negative_weights = ((1.0 - cfg.TRAIN.RPN_POSITIVE_WEIGHT) /
+                                np.sum(labels == 0))
+        bbox_outside_weights[labels == 1, :] = positive_weights
+        bbox_outside_weights[labels == 0, :] = negative_weights
+
+        if DEBUG:
+            self._sums += bbox_targets[labels == 1, :].sum(axis=0)
+            self._squared_sums += (bbox_targets[labels == 1, :] ** 2).sum(axis=0)
+            self._counts += np.sum(labels == 1)
+            means = self._sums / self._counts
+            stds = np.sqrt(self._squared_sums / self._counts - means ** 2)
+            print 'means:'
+            print means
+            print 'stdevs:'
+            print stds
+
+        # map up to original set of anchors
+        labels = _unmap(labels, total_anchors, inds_inside, fill=-1)
+        bbox_targets = _unmap(bbox_targets, total_anchors, inds_inside, fill=0)
+        bbox_inside_weights = _unmap(bbox_inside_weights, total_anchors, inds_inside, fill=0)
+        bbox_outside_weights = _unmap(bbox_outside_weights, total_anchors, inds_inside, fill=0)
+
+        if DEBUG:
+            print 'rpn: max max_overlap', np.max(max_overlaps)
+            print 'rpn: num_positive', np.sum(labels == 1)
+            print 'rpn: num_negative', np.sum(labels == 0)
+            self._fg_sum += np.sum(labels == 1)
+            self._bg_sum += np.sum(labels == 0)
+            self._count += 1
+            print 'rpn: num_positive avg', self._fg_sum / self._count
+            print 'rpn: num_negative avg', self._bg_sum / self._count
+
+        # labels
+        labels = labels.reshape((1, height, width, A)).transpose(0, 3, 1, 2)
+        labels = labels.reshape((1, 1, A * height, width))
+        top[0].reshape(*labels.shape)
+        top[0].data[...] = labels
+
+        # bbox_targets
+        bbox_targets = bbox_targets \
+            .reshape((1, height, width, A * 4)).transpose(0, 3, 1, 2)
+        top[1].reshape(*bbox_targets.shape)
+        top[1].data[...] = bbox_targets
+
+        # bbox_inside_weights
+        bbox_inside_weights = bbox_inside_weights \
+            .reshape((1, height, width, A * 4)).transpose(0, 3, 1, 2)
+        assert bbox_inside_weights.shape[2] == height
+        assert bbox_inside_weights.shape[3] == width
+        top[2].reshape(*bbox_inside_weights.shape)
+        top[2].data[...] = bbox_inside_weights
+
+        # bbox_outside_weights
+        bbox_outside_weights = bbox_outside_weights \
+            .reshape((1, height, width, A * 4)).transpose(0, 3, 1, 2)
+        assert bbox_outside_weights.shape[2] == height
+        assert bbox_outside_weights.shape[3] == width
+        top[3].reshape(*bbox_outside_weights.shape)
+        top[3].data[...] = bbox_outside_weights
+
+    def backward(self, top, propagate_down, bottom):
+        """This layer does not propagate gradients."""
+        pass
+
+    def reshape(self, bottom, top):
+        """Reshaping happens during the call to forward."""
+        pass
+
+
+def _unmap(data, count, inds, fill=0):
+    """ Unmap a subset of item (data) back to the original set of items (of
+    size count) """
+    if len(data.shape) == 1:
+        ret = np.empty((count, ), dtype=np.float32)
+        ret.fill(fill)
+        ret[inds] = data
+    else:
+        ret = np.empty((count, ) + data.shape[1:], dtype=np.float32)
+        ret.fill(fill)
+        ret[inds, :] = data
+    return ret
+
+
+def _compute_targets(ex_rois, gt_rois):
+    """Compute bounding-box regression targets for an image."""
+
+    assert ex_rois.shape[0] == gt_rois.shape[0]
+    assert ex_rois.shape[1] == 4
+    assert gt_rois.shape[1] == 5
+
+    return bbox_transform(ex_rois, gt_rois[:, :4]).astype(np.float32, copy=False)
diff --git a/lib/rpn/generate.py b/lib/rpn/generate.py
new file mode 100644
index 0000000..060daf4
--- /dev/null
+++ b/lib/rpn/generate.py
@@ -0,0 +1,117 @@
+# --------------------------------------------------------
+# Faster R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+from fast_rcnn.config import cfg
+from utils.blob import im_list_to_blob
+from utils.timer import Timer
+import numpy as np
+import cv2
+
+def _vis_proposals(im, dets, thresh=0.5):
+    """Draw detected bounding boxes."""
+    inds = np.where(dets[:, -1] >= thresh)[0]
+    if len(inds) == 0:
+        return
+
+    class_name = 'obj'
+    im = im[:, :, (2, 1, 0)]
+    fig, ax = plt.subplots(figsize=(12, 12))
+    ax.imshow(im, aspect='equal')
+    for i in inds:
+        bbox = dets[i, :4]
+        score = dets[i, -1]
+
+        ax.add_patch(
+            plt.Rectangle((bbox[0], bbox[1]),
+                          bbox[2] - bbox[0],
+                          bbox[3] - bbox[1], fill=False,
+                          edgecolor='red', linewidth=3.5)
+            )
+        ax.text(bbox[0], bbox[1] - 2,
+                '{:s} {:.3f}'.format(class_name, score),
+                bbox=dict(facecolor='blue', alpha=0.5),
+                fontsize=14, color='white')
+
+    ax.set_title(('{} detections with '
+                  'p({} | box) >= {:.1f}').format(class_name, class_name,
+                                                  thresh),
+                  fontsize=14)
+    plt.axis('off')
+    plt.tight_layout()
+    plt.draw()
+
+def _get_image_blob(im):
+    """Converts an image into a network input.
+
+    Arguments:
+        im (ndarray): a color image in BGR order
+
+    Returns:
+        blob (ndarray): a data blob holding an image pyramid
+        im_scale_factors (list): list of image scales (relative to im) used
+            in the image pyramid
+    """
+    im_orig = im.astype(np.float32, copy=True)
+    im_orig -= cfg.PIXEL_MEANS
+
+    im_shape = im_orig.shape
+    im_size_min = np.min(im_shape[0:2])
+    im_size_max = np.max(im_shape[0:2])
+
+    processed_ims = []
+
+    assert len(cfg.TEST.SCALES) == 1
+    target_size = cfg.TEST.SCALES[0]
+
+    im_scale = float(target_size) / float(im_size_min)
+    # Prevent the biggest axis from being more than MAX_SIZE
+    if np.round(im_scale * im_size_max) > cfg.TEST.MAX_SIZE:
+        im_scale = float(cfg.TEST.MAX_SIZE) / float(im_size_max)
+    im = cv2.resize(im_orig, None, None, fx=im_scale, fy=im_scale,
+                    interpolation=cv2.INTER_LINEAR)
+    im_info = np.hstack((im.shape[:2], im_scale))[np.newaxis, :]
+    processed_ims.append(im)
+
+    # Create a blob to hold the input images
+    blob = im_list_to_blob(processed_ims)
+
+    return blob, im_info
+
+def im_proposals(net, im):
+    """Generate RPN proposals on a single image."""
+    blobs = {}
+    blobs['data'], blobs['im_info'] = _get_image_blob(im)
+    net.blobs['data'].reshape(*(blobs['data'].shape))
+    net.blobs['im_info'].reshape(*(blobs['im_info'].shape))
+    blobs_out = net.forward(
+            data=blobs['data'].astype(np.float32, copy=False),
+            im_info=blobs['im_info'].astype(np.float32, copy=False))
+
+    scale = blobs['im_info'][0, 2]
+    boxes = blobs_out['rois'][:, 1:].copy() / scale
+    scores = blobs_out['scores'].copy()
+    return boxes, scores
+
+def imdb_proposals(net, imdb):
+    """Generate RPN proposals on all images in an imdb."""
+
+    _t = Timer()
+    imdb_boxes = [[] for _ in xrange(imdb.num_images)]
+    for i in xrange(imdb.num_images):
+        im = cv2.imread(imdb.image_path_at(i))
+        _t.tic()
+        imdb_boxes[i], scores = im_proposals(net, im)
+        _t.toc()
+        print 'im_proposals: {:d}/{:d} {:.3f}s' \
+              .format(i + 1, imdb.num_images, _t.average_time)
+        if 0:
+            dets = np.hstack((imdb_boxes[i], scores))
+            # from IPython import embed; embed()
+            _vis_proposals(im, dets[:3, :], thresh=0.9)
+            plt.show()
+
+    return imdb_boxes
diff --git a/lib/rpn/generate_anchors.py b/lib/rpn/generate_anchors.py
new file mode 100644
index 0000000..1125a80
--- /dev/null
+++ b/lib/rpn/generate_anchors.py
@@ -0,0 +1,105 @@
+# --------------------------------------------------------
+# Faster R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick and Sean Bell
+# --------------------------------------------------------
+
+import numpy as np
+
+# Verify that we compute the same anchors as Shaoqing's matlab implementation:
+#
+#    >> load output/rpn_cachedir/faster_rcnn_VOC2007_ZF_stage1_rpn/anchors.mat
+#    >> anchors
+#
+#    anchors =
+#
+#       -83   -39   100    56
+#      -175   -87   192   104
+#      -359  -183   376   200
+#       -55   -55    72    72
+#      -119  -119   136   136
+#      -247  -247   264   264
+#       -35   -79    52    96
+#       -79  -167    96   184
+#      -167  -343   184   360
+
+#array([[ -83.,  -39.,  100.,   56.],
+#       [-175.,  -87.,  192.,  104.],
+#       [-359., -183.,  376.,  200.],
+#       [ -55.,  -55.,   72.,   72.],
+#       [-119., -119.,  136.,  136.],
+#       [-247., -247.,  264.,  264.],
+#       [ -35.,  -79.,   52.,   96.],
+#       [ -79., -167.,   96.,  184.],
+#       [-167., -343.,  184.,  360.]])
+
+def generate_anchors(base_size=16, ratios=[0.5, 1, 2],
+                     scales=2**np.arange(3, 6)):
+    """
+    Generate anchor (reference) windows by enumerating aspect ratios X
+    scales wrt a reference (0, 0, 15, 15) window.
+    """
+
+    base_anchor = np.array([1, 1, base_size, base_size]) - 1
+    ratio_anchors = _ratio_enum(base_anchor, ratios)
+    anchors = np.vstack([_scale_enum(ratio_anchors[i, :], scales)
+                         for i in xrange(ratio_anchors.shape[0])])
+    return anchors
+
+def _whctrs(anchor):
+    """
+    Return width, height, x center, and y center for an anchor (window).
+    """
+
+    w = anchor[2] - anchor[0] + 1
+    h = anchor[3] - anchor[1] + 1
+    x_ctr = anchor[0] + 0.5 * (w - 1)
+    y_ctr = anchor[1] + 0.5 * (h - 1)
+    return w, h, x_ctr, y_ctr
+
+def _mkanchors(ws, hs, x_ctr, y_ctr):
+    """
+    Given a vector of widths (ws) and heights (hs) around a center
+    (x_ctr, y_ctr), output a set of anchors (windows).
+    """
+
+    ws = ws[:, np.newaxis]
+    hs = hs[:, np.newaxis]
+    anchors = np.hstack((x_ctr - 0.5 * (ws - 1),
+                         y_ctr - 0.5 * (hs - 1),
+                         x_ctr + 0.5 * (ws - 1),
+                         y_ctr + 0.5 * (hs - 1)))
+    return anchors
+
+def _ratio_enum(anchor, ratios):
+    """
+    Enumerate a set of anchors for each aspect ratio wrt an anchor.
+    """
+
+    w, h, x_ctr, y_ctr = _whctrs(anchor)
+    size = w * h
+    size_ratios = size / ratios
+    ws = np.round(np.sqrt(size_ratios))
+    hs = np.round(ws * ratios)
+    anchors = _mkanchors(ws, hs, x_ctr, y_ctr)
+    return anchors
+
+def _scale_enum(anchor, scales):
+    """
+    Enumerate a set of anchors for each scale wrt an anchor.
+    """
+
+    w, h, x_ctr, y_ctr = _whctrs(anchor)
+    ws = w * scales
+    hs = h * scales
+    anchors = _mkanchors(ws, hs, x_ctr, y_ctr)
+    return anchors
+
+if __name__ == '__main__':
+    import time
+    t = time.time()
+    a = generate_anchors()
+    print time.time() - t
+    print a
+    from IPython import embed; embed()
diff --git a/lib/rpn/proposal_layer.py b/lib/rpn/proposal_layer.py
new file mode 100644
index 0000000..24b9db7
--- /dev/null
+++ b/lib/rpn/proposal_layer.py
@@ -0,0 +1,177 @@
+# --------------------------------------------------------
+# Faster R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick and Sean Bell
+# --------------------------------------------------------
+
+import caffe
+import numpy as np
+import yaml
+from fast_rcnn.config import cfg
+from generate_anchors import generate_anchors
+from fast_rcnn.bbox_transform import bbox_transform_inv, clip_boxes
+from fast_rcnn.nms_wrapper import nms
+
+DEBUG = False
+
+class ProposalLayer(caffe.Layer):
+    """
+    Outputs object detection proposals by applying estimated bounding-box
+    transformations to a set of regular boxes (called "anchors").
+    """
+
+    def setup(self, bottom, top):
+        # parse the layer parameter string, which must be valid YAML
+        layer_params = yaml.load(self.param_str)
+
+        self._feat_stride = layer_params['feat_stride']
+        anchor_scales = layer_params.get('scales', (8, 16, 32))
+        self._anchors = generate_anchors(scales=np.array(anchor_scales))
+        self._num_anchors = self._anchors.shape[0]
+
+        if DEBUG:
+            print 'feat_stride: {}'.format(self._feat_stride)
+            print 'anchors:'
+            print self._anchors
+
+        # rois blob: holds R regions of interest, each is a 5-tuple
+        # (n, x1, y1, x2, y2) specifying an image batch index n and a
+        # rectangle (x1, y1, x2, y2)
+        top[0].reshape(1, 5)
+
+        # scores blob: holds scores for R regions of interest
+        if len(top) > 1:
+            top[1].reshape(1, 1, 1, 1)
+
+    def forward(self, bottom, top):
+        # Algorithm:
+        #
+        # for each (H, W) location i
+        #   generate A anchor boxes centered on cell i
+        #   apply predicted bbox deltas at cell i to each of the A anchors
+        # clip predicted boxes to image
+        # remove predicted boxes with either height or width < threshold
+        # sort all (proposal, score) pairs by score from highest to lowest
+        # take top pre_nms_topN proposals before NMS
+        # apply NMS with threshold 0.7 to remaining proposals
+        # take after_nms_topN proposals after NMS
+        # return the top proposals (-> RoIs top, scores top)
+
+        assert bottom[0].data.shape[0] == 1, \
+            'Only single item batches are supported'
+
+        cfg_key = str('TRAIN' if self.phase == 0 else 'TEST') # either 'TRAIN' or 'TEST'
+        pre_nms_topN  = cfg[cfg_key].RPN_PRE_NMS_TOP_N
+        post_nms_topN = cfg[cfg_key].RPN_POST_NMS_TOP_N
+        nms_thresh    = cfg[cfg_key].RPN_NMS_THRESH
+        min_size      = cfg[cfg_key].RPN_MIN_SIZE
+
+        # the first set of _num_anchors channels are bg probs
+        # the second set are the fg probs, which we want
+        scores = bottom[0].data[:, self._num_anchors:, :, :]
+        bbox_deltas = bottom[1].data
+        im_info = bottom[2].data[0, :]
+
+        if DEBUG:
+            print 'im_size: ({}, {})'.format(im_info[0], im_info[1])
+            print 'scale: {}'.format(im_info[2])
+
+        # 1. Generate proposals from bbox deltas and shifted anchors
+        height, width = scores.shape[-2:]
+
+        if DEBUG:
+            print 'score map size: {}'.format(scores.shape)
+
+        # Enumerate all shifts
+        shift_x = np.arange(0, width) * self._feat_stride
+        shift_y = np.arange(0, height) * self._feat_stride
+        shift_x, shift_y = np.meshgrid(shift_x, shift_y)
+        shifts = np.vstack((shift_x.ravel(), shift_y.ravel(),
+                            shift_x.ravel(), shift_y.ravel())).transpose()
+
+        # Enumerate all shifted anchors:
+        #
+        # add A anchors (1, A, 4) to
+        # cell K shifts (K, 1, 4) to get
+        # shift anchors (K, A, 4)
+        # reshape to (K*A, 4) shifted anchors
+        A = self._num_anchors
+        K = shifts.shape[0]
+        anchors = self._anchors.reshape((1, A, 4)) + \
+                  shifts.reshape((1, K, 4)).transpose((1, 0, 2))
+        anchors = anchors.reshape((K * A, 4))
+
+        # Transpose and reshape predicted bbox transformations to get them
+        # into the same order as the anchors:
+        #
+        # bbox deltas will be (1, 4 * A, H, W) format
+        # transpose to (1, H, W, 4 * A)
+        # reshape to (1 * H * W * A, 4) where rows are ordered by (h, w, a)
+        # in slowest to fastest order
+        bbox_deltas = bbox_deltas.transpose((0, 2, 3, 1)).reshape((-1, 4))
+
+        # Same story for the scores:
+        #
+        # scores are (1, A, H, W) format
+        # transpose to (1, H, W, A)
+        # reshape to (1 * H * W * A, 1) where rows are ordered by (h, w, a)
+        scores = scores.transpose((0, 2, 3, 1)).reshape((-1, 1))
+
+        # Convert anchors into proposals via bbox transformations
+        proposals = bbox_transform_inv(anchors, bbox_deltas)
+
+        # 2. clip predicted boxes to image
+        proposals = clip_boxes(proposals, im_info[:2])
+
+        # 3. remove predicted boxes with either height or width < threshold
+        # (NOTE: convert min_size to input image scale stored in im_info[2])
+        keep = _filter_boxes(proposals, min_size * im_info[2])
+        proposals = proposals[keep, :]
+        scores = scores[keep]
+
+        # 4. sort all (proposal, score) pairs by score from highest to lowest
+        # 5. take top pre_nms_topN (e.g. 6000)
+        order = scores.ravel().argsort()[::-1]
+        if pre_nms_topN > 0:
+            order = order[:pre_nms_topN]
+        proposals = proposals[order, :]
+        scores = scores[order]
+
+        # 6. apply nms (e.g. threshold = 0.7)
+        # 7. take after_nms_topN (e.g. 300)
+        # 8. return the top proposals (-> RoIs top)
+        keep = nms(np.hstack((proposals, scores)), nms_thresh)
+        if post_nms_topN > 0:
+            keep = keep[:post_nms_topN]
+        proposals = proposals[keep, :]
+        scores = scores[keep]
+
+        # Output rois blob
+        # Our RPN implementation only supports a single input image, so all
+        # batch inds are 0
+        batch_inds = np.zeros((proposals.shape[0], 1), dtype=np.float32)
+        blob = np.hstack((batch_inds, proposals.astype(np.float32, copy=False)))
+        # print blob.shape
+        top[0].reshape(*(blob.shape))
+        top[0].data[...] = blob
+
+        # [Optional] output scores blob
+        if len(top) > 1:
+            top[1].reshape(*(scores.shape))
+            top[1].data[...] = scores
+
+    def backward(self, top, propagate_down, bottom):
+        """This layer does not propagate gradients."""
+        pass
+
+    def reshape(self, bottom, top):
+        """Reshaping happens during the call to forward."""
+        pass
+
+def _filter_boxes(boxes, min_size):
+    """Remove all boxes with any side smaller than min_size."""
+    ws = boxes[:, 2] - boxes[:, 0] + 1
+    hs = boxes[:, 3] - boxes[:, 1] + 1
+    keep = np.where((ws >= min_size) & (hs >= min_size))[0]
+    return keep
diff --git a/lib/rpn/proposal_target_layer.py b/lib/rpn/proposal_target_layer.py
new file mode 100644
index 0000000..0bf2f5a
--- /dev/null
+++ b/lib/rpn/proposal_target_layer.py
@@ -0,0 +1,216 @@
+# --------------------------------------------------------
+# Faster R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick and Sean Bell
+# --------------------------------------------------------
+
+import caffe
+import yaml
+import numpy as np
+import numpy.random as npr
+from fast_rcnn.config import cfg
+from fast_rcnn.bbox_transform import bbox_transform
+from utils.cython_bbox import bbox_overlaps
+
+DEBUG = False
+
+class ProposalTargetLayer(caffe.Layer):
+    """
+    Assign object detection proposals to ground-truth targets. Produces proposal
+    classification labels and bounding-box regression targets.
+    """
+
+    def setup(self, bottom, top):
+        layer_params = yaml.load(self.param_str)
+        self._num_classes = layer_params['num_classes']
+
+        # sampled rois (0, x1, y1, x2, y2)
+        top[0].reshape(1, 5, 1, 1)
+        # labels
+        top[1].reshape(1, 1, 1, 1)
+        # bbox_targets
+        top[2].reshape(1, self._num_classes * 4, 1, 1)
+        # bbox_inside_weights
+        top[3].reshape(1, self._num_classes * 4, 1, 1)
+        # bbox_outside_weights
+        top[4].reshape(1, self._num_classes * 4, 1, 1)
+
+    def forward(self, bottom, top):
+        # Proposal ROIs (0, x1, y1, x2, y2) coming from RPN
+        # (i.e., rpn.proposal_layer.ProposalLayer), or any other source
+        all_rois = bottom[0].data
+        # GT boxes (x1, y1, x2, y2, label)
+        # TODO(rbg): it's annoying that sometimes I have extra info before
+        # and other times after box coordinates -- normalize to one format
+        gt_boxes = bottom[1].data
+
+        # Include ground-truth boxes in the set of candidate rois
+        zeros = np.zeros((gt_boxes.shape[0], 1), dtype=gt_boxes.dtype)
+        all_rois = np.vstack(
+            (all_rois, np.hstack((zeros, gt_boxes[:, :-1])))
+        )
+
+        # Sanity check: single batch only
+        assert np.all(all_rois[:, 0] == 0), \
+                'Only single item batches are supported'
+
+        rois_per_image = np.inf if cfg.TRAIN.BATCH_SIZE == -1 else cfg.TRAIN.BATCH_SIZE
+        fg_rois_per_image = np.round(cfg.TRAIN.FG_FRACTION * rois_per_image)
+
+        # Sample rois with classification labels and bounding box regression
+        # targets
+        # print 'proposal_target_layer:', fg_rois_per_image
+        labels, rois, bbox_targets, bbox_inside_weights = _sample_rois(
+            all_rois, gt_boxes, fg_rois_per_image,
+            rois_per_image, self._num_classes)
+
+        if DEBUG:
+            print 'num fg: {}'.format((labels > 0).sum())
+            print 'num bg: {}'.format((labels == 0).sum())
+            self._count += 1
+            self._fg_num += (labels > 0).sum()
+            self._bg_num += (labels == 0).sum()
+            print 'num fg avg: {}'.format(self._fg_num / self._count)
+            print 'num bg avg: {}'.format(self._bg_num / self._count)
+            print 'ratio: {:.3f}'.format(float(self._fg_num) / float(self._bg_num))
+
+        # sampled rois
+        # modified by ywxiong
+        rois = rois.reshape((rois.shape[0], rois.shape[1], 1, 1))
+        top[0].reshape(*rois.shape)
+        top[0].data[...] = rois
+
+        # classification labels
+        # modified by ywxiong
+        labels = labels.reshape((labels.shape[0], 1, 1, 1))
+        top[1].reshape(*labels.shape)
+        top[1].data[...] = labels
+
+        # bbox_targets
+        # modified by ywxiong
+        bbox_targets = bbox_targets.reshape((bbox_targets.shape[0], bbox_targets.shape[1], 1, 1))
+        top[2].reshape(*bbox_targets.shape)
+        top[2].data[...] = bbox_targets
+
+        # bbox_inside_weights
+        # modified by ywxiong
+        bbox_inside_weights = bbox_inside_weights.reshape((bbox_inside_weights.shape[0], bbox_inside_weights.shape[1], 1, 1))
+        top[3].reshape(*bbox_inside_weights.shape)
+        top[3].data[...] = bbox_inside_weights
+
+        # bbox_outside_weights
+        # modified by ywxiong
+        bbox_inside_weights = bbox_inside_weights.reshape((bbox_inside_weights.shape[0], bbox_inside_weights.shape[1], 1, 1))
+        top[4].reshape(*bbox_inside_weights.shape)
+        top[4].data[...] = np.array(bbox_inside_weights > 0).astype(np.float32)
+
+    def backward(self, top, propagate_down, bottom):
+        """This layer does not propagate gradients."""
+        pass
+
+    def reshape(self, bottom, top):
+        """Reshaping happens during the call to forward."""
+        pass
+
+
+def _get_bbox_regression_labels(bbox_target_data, num_classes):
+    """Bounding-box regression targets (bbox_target_data) are stored in a
+    compact form N x (class, tx, ty, tw, th)
+
+    This function expands those targets into the 4-of-4*K representation used
+    by the network (i.e. only one class has non-zero targets).
+
+    Returns:
+        bbox_target (ndarray): N x 4K blob of regression targets
+        bbox_inside_weights (ndarray): N x 4K blob of loss weights
+    """
+
+    clss = bbox_target_data[:, 0]
+    bbox_targets = np.zeros((clss.size, 4 * num_classes), dtype=np.float32)
+    # print 'proposal_target_layer:', bbox_targets.shape
+    bbox_inside_weights = np.zeros(bbox_targets.shape, dtype=np.float32)
+    inds = np.where(clss > 0)[0]
+    if cfg.TRAIN.AGONISTIC:
+        for ind in inds:
+            cls = clss[ind]
+            start = 4 * (1 if cls > 0 else 0)
+            end = start + 4
+            bbox_targets[ind, start:end] = bbox_target_data[ind, 1:]
+            bbox_inside_weights[ind, start:end] = cfg.TRAIN.BBOX_INSIDE_WEIGHTS
+    else:
+        for ind in inds:
+            cls = clss[ind]
+            start = 4 * cls
+            end = start + 4
+            bbox_targets[ind, start:end] = bbox_target_data[ind, 1:]
+            bbox_inside_weights[ind, start:end] = cfg.TRAIN.BBOX_INSIDE_WEIGHTS
+    return bbox_targets, bbox_inside_weights
+
+
+def _compute_targets(ex_rois, gt_rois, labels):
+    """Compute bounding-box regression targets for an image."""
+
+    assert ex_rois.shape[0] == gt_rois.shape[0]
+    assert ex_rois.shape[1] == 4
+    assert gt_rois.shape[1] == 4
+
+    targets = bbox_transform(ex_rois, gt_rois)
+    if cfg.TRAIN.BBOX_NORMALIZE_TARGETS_PRECOMPUTED:
+        # Optionally normalize targets by a precomputed mean and stdev
+        targets = ((targets - np.array(cfg.TRAIN.BBOX_NORMALIZE_MEANS))
+                / np.array(cfg.TRAIN.BBOX_NORMALIZE_STDS))
+    return np.hstack(
+            (labels[:, np.newaxis], targets)).astype(np.float32, copy=False)
+
+def _sample_rois(all_rois, gt_boxes, fg_rois_per_image, rois_per_image, num_classes):
+    """Generate a random sample of RoIs comprising foreground and background
+    examples.
+    """
+    # overlaps: (rois x gt_boxes)
+    overlaps = bbox_overlaps(
+        np.ascontiguousarray(all_rois[:, 1:5], dtype=np.float),
+        np.ascontiguousarray(gt_boxes[:, :4], dtype=np.float))
+    gt_assignment = overlaps.argmax(axis=1)
+    max_overlaps = overlaps.max(axis=1)
+    labels = gt_boxes[gt_assignment, 4]
+
+    # Select foreground RoIs as those with >= FG_THRESH overlap
+    fg_inds = np.where(max_overlaps >= cfg.TRAIN.FG_THRESH)[0]
+    # Guard against the case when an image has fewer than fg_rois_per_image
+    # foreground RoIs
+    fg_rois_per_this_image = min(fg_rois_per_image, fg_inds.size)
+    # Sample foreground regions without replacement
+    if fg_inds.size > 0:
+        fg_inds = npr.choice(fg_inds, size=fg_rois_per_this_image, replace=False)
+
+    # Select background RoIs as those within [BG_THRESH_LO, BG_THRESH_HI)
+    bg_inds = np.where((max_overlaps < cfg.TRAIN.BG_THRESH_HI) &
+                       (max_overlaps >= cfg.TRAIN.BG_THRESH_LO))[0]
+    # Compute number of background RoIs to take from this image (guarding
+    # against there being fewer than desired)
+    bg_rois_per_this_image = rois_per_image - fg_rois_per_this_image
+    bg_rois_per_this_image = min(bg_rois_per_this_image, bg_inds.size)
+    # Sample background regions without replacement
+    if bg_inds.size > 0:
+        bg_inds = npr.choice(bg_inds, size=bg_rois_per_this_image, replace=False)
+
+    # The indices that we're selecting (both fg and bg)
+    keep_inds = np.append(fg_inds, bg_inds)
+    # print 'proposal_target_layer:', keep_inds
+    
+    # Select sampled values from various arrays:
+    labels = labels[keep_inds]
+    # Clamp labels for the background RoIs to 0
+    labels[fg_rois_per_this_image:] = 0
+    rois = all_rois[keep_inds]
+    
+    # print 'proposal_target_layer:', rois
+    bbox_target_data = _compute_targets(
+        rois[:, 1:5], gt_boxes[gt_assignment[keep_inds], :4], labels)
+
+    # print 'proposal_target_layer:', bbox_target_data
+    bbox_targets, bbox_inside_weights = \
+        _get_bbox_regression_labels(bbox_target_data, num_classes)
+
+    return labels, rois, bbox_targets, bbox_inside_weights
diff --git a/lib/setup.py b/lib/setup.py
new file mode 100644
index 0000000..0f4615f
--- /dev/null
+++ b/lib/setup.py
@@ -0,0 +1,156 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+import os
+from os.path import join as pjoin
+from setuptools import setup
+from distutils.extension import Extension
+from Cython.Distutils import build_ext
+import subprocess
+import numpy as np
+
+def find_in_path(name, path):
+    "Find a file in a search path"
+    # Adapted fom
+    # http://code.activestate.com/recipes/52224-find-a-file-given-a-search-path/
+    for dir in path.split(os.pathsep):
+        binpath = pjoin(dir, name)
+        if os.path.exists(binpath):
+            return os.path.abspath(binpath)
+    return None
+
+
+def locate_cuda():
+    """Locate the CUDA environment on the system
+
+    Returns a dict with keys 'home', 'nvcc', 'include', and 'lib64'
+    and values giving the absolute path to each directory.
+
+    Starts by looking for the CUDAHOME env variable. If not found, everything
+    is based on finding 'nvcc' in the PATH.
+    """
+
+    # first check if the CUDAHOME env variable is in use
+    if 'CUDAHOME' in os.environ:
+        home = os.environ['CUDAHOME']
+        nvcc = pjoin(home, 'bin', 'nvcc')
+    else:
+        # otherwise, search the PATH for NVCC
+        default_path = pjoin(os.sep, 'usr', 'local', 'cuda', 'bin')
+        nvcc = find_in_path('nvcc', os.environ['PATH'] + os.pathsep + default_path)
+        if nvcc is None:
+            raise EnvironmentError('The nvcc binary could not be '
+                'located in your $PATH. Either add it to your path, or set $CUDAHOME')
+        home = os.path.dirname(os.path.dirname(nvcc))
+
+    cudaconfig = {'home':home, 'nvcc':nvcc,
+                  'include': pjoin(home, 'include'),
+                  'lib64': pjoin(home, 'lib64')}
+    for k, v in cudaconfig.iteritems():
+        if not os.path.exists(v):
+            raise EnvironmentError('The CUDA %s path could not be located in %s' % (k, v))
+
+    return cudaconfig
+CUDA = locate_cuda()
+
+
+# Obtain the numpy include directory.  This logic works across numpy versions.
+try:
+    numpy_include = np.get_include()
+except AttributeError:
+    numpy_include = np.get_numpy_include()
+
+def customize_compiler_for_nvcc(self):
+    """inject deep into distutils to customize how the dispatch
+    to gcc/nvcc works.
+
+    If you subclass UnixCCompiler, it's not trivial to get your subclass
+    injected in, and still have the right customizations (i.e.
+    distutils.sysconfig.customize_compiler) run on it. So instead of going
+    the OO route, I have this. Note, it's kindof like a wierd functional
+    subclassing going on."""
+
+    # tell the compiler it can processes .cu
+    self.src_extensions.append('.cu')
+
+    # save references to the default compiler_so and _comple methods
+    default_compiler_so = self.compiler_so
+    super = self._compile
+
+    # now redefine the _compile method. This gets executed for each
+    # object but distutils doesn't have the ability to change compilers
+    # based on source extension: we add it.
+    def _compile(obj, src, ext, cc_args, extra_postargs, pp_opts):
+        if os.path.splitext(src)[1] == '.cu':
+            # use the cuda for .cu files
+            self.set_executable('compiler_so', CUDA['nvcc'])
+            # use only a subset of the extra_postargs, which are 1-1 translated
+            # from the extra_compile_args in the Extension class
+            postargs = extra_postargs['nvcc']
+        else:
+            postargs = extra_postargs['gcc']
+
+        super(obj, src, ext, cc_args, postargs, pp_opts)
+        # reset the default compiler_so, which we might have changed for cuda
+        self.compiler_so = default_compiler_so
+
+    # inject our redefined _compile method into the class
+    self._compile = _compile
+
+
+# run the customize_compiler
+class custom_build_ext(build_ext):
+    def build_extensions(self):
+        customize_compiler_for_nvcc(self.compiler)
+        build_ext.build_extensions(self)
+
+
+ext_modules = [
+    Extension(
+        "utils.cython_bbox",
+        ["utils/bbox.pyx"],
+        extra_compile_args={'gcc': ["-Wno-cpp", "-Wno-unused-function"]},
+        include_dirs = [numpy_include]
+    ),
+    Extension(
+        "nms.cpu_nms",
+        ["nms/cpu_nms.pyx"],
+        extra_compile_args={'gcc': ["-Wno-cpp", "-Wno-unused-function"]},
+        include_dirs = [numpy_include]
+    ),
+    Extension('nms.gpu_nms',
+        ['nms/nms_kernel.cu', 'nms/gpu_nms.pyx'],
+        library_dirs=[CUDA['lib64']],
+        libraries=['cudart'],
+        language='c++',
+        runtime_library_dirs=[CUDA['lib64']],
+        # this syntax is specific to this build system
+        # we're only going to use certain compiler args with nvcc and not with
+        # gcc the implementation of this trick is in customize_compiler() below
+        extra_compile_args={'gcc': ["-Wno-unused-function"],
+                            'nvcc': ['-arch=sm_35',
+                                     '--ptxas-options=-v',
+                                     '-c',
+                                     '--compiler-options',
+                                     "'-fPIC'"]},
+        include_dirs = [numpy_include, CUDA['include']]
+    ),
+    Extension(
+        'pycocotools._mask',
+        sources=['pycocotools/maskApi.c', 'pycocotools/_mask.pyx'],
+        include_dirs = [numpy_include, 'pycocotools'],
+        extra_compile_args={
+            'gcc': ['-Wno-cpp', '-Wno-unused-function', '-std=c99']},
+    ),
+]
+
+setup(
+    name='fast_rcnn',
+    ext_modules=ext_modules,
+    # inject our custom trigger
+    cmdclass={'build_ext': custom_build_ext},
+)
diff --git a/lib/transform/__init__.py b/lib/transform/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/lib/transform/torch_image_transform_layer.py b/lib/transform/torch_image_transform_layer.py
new file mode 100644
index 0000000..9273b3a
--- /dev/null
+++ b/lib/transform/torch_image_transform_layer.py
@@ -0,0 +1,64 @@
+# --------------------------------------------------------
+# Fast/er R-CNN
+# Licensed under The MIT License [see LICENSE for details]
+# --------------------------------------------------------
+
+""" Transform images for compatibility with models trained with
+https://github.com/facebook/fb.resnet.torch.
+
+Usage in model prototxt:
+
+layer {
+  name: 'data_xform'
+  type: 'Python'
+  bottom: 'data_caffe'
+  top: 'data'
+  python_param {
+    module: 'transform.torch_image_transform_layer'
+    layer: 'TorchImageTransformLayer'
+  }
+}
+"""
+
+import caffe
+from fast_rcnn.config import cfg
+import numpy as np
+
+class TorchImageTransformLayer(caffe.Layer):
+    def setup(self, bottom, top):
+        # (1, 3, 1, 1) shaped arrays
+        self.PIXEL_MEANS = \
+            np.array([[[[0.48462227599918]],
+                       [[0.45624044862054]],
+                       [[0.40588363755159]]]])
+        self.PIXEL_STDS = \
+            np.array([[[[0.22889466674951]],
+                       [[0.22446679341259]],
+                       [[0.22495548344775]]]])
+        # The default ("old") pixel means that were already subtracted
+        channel_swap = (0, 3, 1, 2)
+        self.OLD_PIXEL_MEANS = \
+            cfg.PIXEL_MEANS[np.newaxis, :, :, :].transpose(channel_swap)
+
+        top[0].reshape(*(bottom[0].shape))
+
+    def forward(self, bottom, top):
+        ims = bottom[0].data
+        # Invert the channel means that were already subtracted
+        ims += self.OLD_PIXEL_MEANS
+        # 1. Permute BGR to RGB and normalize to [0, 1]
+        ims = ims[:, [2, 1, 0], :, :] / 255.0
+        # 2. Remove channel means
+        ims -= self.PIXEL_MEANS
+        # 3. Standardize channels
+        ims /= self.PIXEL_STDS
+        top[0].reshape(*(ims.shape))
+        top[0].data[...] = ims
+
+    def backward(self, top, propagate_down, bottom):
+        """This layer does not propagate gradients."""
+        pass
+
+    def reshape(self, bottom, top):
+        """Reshaping happens during the call to forward."""
+        pass
diff --git a/lib/utils/.gitignore b/lib/utils/.gitignore
new file mode 100644
index 0000000..4b8a745
--- /dev/null
+++ b/lib/utils/.gitignore
@@ -0,0 +1,2 @@
+*.c
+*.so
diff --git a/lib/utils/__init__.py b/lib/utils/__init__.py
new file mode 100644
index 0000000..7ba6a65
--- /dev/null
+++ b/lib/utils/__init__.py
@@ -0,0 +1,6 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
diff --git a/lib/utils/bbox.pyx b/lib/utils/bbox.pyx
new file mode 100644
index 0000000..e14780d
--- /dev/null
+++ b/lib/utils/bbox.pyx
@@ -0,0 +1,55 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Sergey Karayev
+# --------------------------------------------------------
+
+cimport cython
+import numpy as np
+cimport numpy as np
+
+DTYPE = np.float
+ctypedef np.float_t DTYPE_t
+
+def bbox_overlaps(
+        np.ndarray[DTYPE_t, ndim=2] boxes,
+        np.ndarray[DTYPE_t, ndim=2] query_boxes):
+    """
+    Parameters
+    ----------
+    boxes: (N, 4) ndarray of float
+    query_boxes: (K, 4) ndarray of float
+    Returns
+    -------
+    overlaps: (N, K) ndarray of overlap between boxes and query_boxes
+    """
+    cdef unsigned int N = boxes.shape[0]
+    cdef unsigned int K = query_boxes.shape[0]
+    cdef np.ndarray[DTYPE_t, ndim=2] overlaps = np.zeros((N, K), dtype=DTYPE)
+    cdef DTYPE_t iw, ih, box_area
+    cdef DTYPE_t ua
+    cdef unsigned int k, n
+    for k in range(K):
+        box_area = (
+            (query_boxes[k, 2] - query_boxes[k, 0] + 1) *
+            (query_boxes[k, 3] - query_boxes[k, 1] + 1)
+        )
+        for n in range(N):
+            iw = (
+                min(boxes[n, 2], query_boxes[k, 2]) -
+                max(boxes[n, 0], query_boxes[k, 0]) + 1
+            )
+            if iw > 0:
+                ih = (
+                    min(boxes[n, 3], query_boxes[k, 3]) -
+                    max(boxes[n, 1], query_boxes[k, 1]) + 1
+                )
+                if ih > 0:
+                    ua = float(
+                        (boxes[n, 2] - boxes[n, 0] + 1) *
+                        (boxes[n, 3] - boxes[n, 1] + 1) +
+                        box_area - iw * ih
+                    )
+                    overlaps[n, k] = iw * ih / ua
+    return overlaps
diff --git a/lib/utils/blob.py b/lib/utils/blob.py
new file mode 100644
index 0000000..1c31642
--- /dev/null
+++ b/lib/utils/blob.py
@@ -0,0 +1,45 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Blob helper functions."""
+
+import numpy as np
+import cv2
+
+def im_list_to_blob(ims):
+    """Convert a list of images into a network input.
+
+    Assumes images are already prepared (means subtracted, BGR order, ...).
+    """
+    max_shape = np.array([im.shape for im in ims]).max(axis=0)
+    num_images = len(ims)
+    blob = np.zeros((num_images, max_shape[0], max_shape[1], 3),
+                    dtype=np.float32)
+    for i in xrange(num_images):
+        im = ims[i]
+        blob[i, 0:im.shape[0], 0:im.shape[1], :] = im
+    # Move channels (axis 3) to axis 1
+    # Axis order will become: (batch elem, channel, height, width)
+    channel_swap = (0, 3, 1, 2)
+    blob = blob.transpose(channel_swap)
+    return blob
+
+def prep_im_for_blob(im, pixel_means, target_size, max_size):
+    """Mean subtract and scale an image for use in a blob."""
+    im = im.astype(np.float32, copy=False)
+    im -= pixel_means
+    im_shape = im.shape
+    im_size_min = np.min(im_shape[0:2])
+    im_size_max = np.max(im_shape[0:2])
+    im_scale = float(target_size) / float(im_size_min)
+    # Prevent the biggest axis from being more than MAX_SIZE
+    if np.round(im_scale * im_size_max) > max_size:
+        im_scale = float(max_size) / float(im_size_max)
+    im = cv2.resize(im, None, None, fx=im_scale, fy=im_scale,
+                    interpolation=cv2.INTER_LINEAR)
+
+    return im, im_scale
diff --git a/lib/utils/timer.py b/lib/utils/timer.py
new file mode 100644
index 0000000..dacc942
--- /dev/null
+++ b/lib/utils/timer.py
@@ -0,0 +1,32 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+import time
+
+class Timer(object):
+    """A simple timer."""
+    def __init__(self):
+        self.total_time = 0.
+        self.calls = 0
+        self.start_time = 0.
+        self.diff = 0.
+        self.average_time = 0.
+
+    def tic(self):
+        # using time.time instead of time.clock because time time.clock
+        # does not normalize for multithreading
+        self.start_time = time.time()
+
+    def toc(self, average=True):
+        self.diff = time.time() - self.start_time
+        self.total_time += self.diff
+        self.calls += 1
+        self.average_time = self.total_time / self.calls
+        if average:
+            return self.average_time
+        else:
+            return self.diff
diff --git a/models/coco/VGG16/fast_rcnn/solver.prototxt b/models/coco/VGG16/fast_rcnn/solver.prototxt
new file mode 100644
index 0000000..e537e1b
--- /dev/null
+++ b/models/coco/VGG16/fast_rcnn/solver.prototxt
@@ -0,0 +1,16 @@
+train_net: "models/coco/VGG16/fast_rcnn/train.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 200000
+display: 20
+average_loss: 100
+# iter_size: 1
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg16_fast_rcnn"
+#debug_info: true
diff --git a/models/coco/VGG16/fast_rcnn/test.prototxt b/models/coco/VGG16/fast_rcnn/test.prototxt
new file mode 100644
index 0000000..5bc1e99
--- /dev/null
+++ b/models/coco/VGG16/fast_rcnn/test.prototxt
@@ -0,0 +1,499 @@
+name: "VGG_ILSVRC_16_layers"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "rois"
+input_shape {
+  dim: 1 # to be changed on-the-fly to num ROIs
+  dim: 5 # [batch ind, x1, y1, x2, y2] zero-based indexing
+}
+
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 81
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 324
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+}
diff --git a/models/coco/VGG16/fast_rcnn/train.prototxt b/models/coco/VGG16/fast_rcnn/train.prototxt
new file mode 100644
index 0000000..1bba398
--- /dev/null
+++ b/models/coco/VGG16/fast_rcnn/train.prototxt
@@ -0,0 +1,485 @@
+name: "VGG_ILSVRC_16_layers"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 81"
+  }
+}
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 81
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 324
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
diff --git a/models/coco/VGG16/faster_rcnn_end2end/solver.prototxt b/models/coco/VGG16/faster_rcnn_end2end/solver.prototxt
new file mode 100644
index 0000000..c8e57db
--- /dev/null
+++ b/models/coco/VGG16/faster_rcnn_end2end/solver.prototxt
@@ -0,0 +1,15 @@
+train_net: "models/coco/VGG16/faster_rcnn_end2end/train.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 350000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg16_faster_rcnn"
+iter_size: 2
diff --git a/models/coco/VGG16/faster_rcnn_end2end/test.prototxt b/models/coco/VGG16/faster_rcnn_end2end/test.prototxt
new file mode 100644
index 0000000..a700a52
--- /dev/null
+++ b/models/coco/VGG16/faster_rcnn_end2end/test.prototxt
@@ -0,0 +1,590 @@
+name: "VGG_ILSVRC_16_layers"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5_3"
+  top: "rpn/output"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 24   # 2(bg/fg) * 12(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 48   # 4 * 12(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 24 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16 \n'scales': !!python/tuple [4, 8, 16, 32]"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 81
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 324
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+}
diff --git a/models/coco/VGG16/faster_rcnn_end2end/train.prototxt b/models/coco/VGG16/faster_rcnn_end2end/train.prototxt
new file mode 100644
index 0000000..d36c92d
--- /dev/null
+++ b/models/coco/VGG16/faster_rcnn_end2end/train.prototxt
@@ -0,0 +1,642 @@
+name: "VGG_ILSVRC_16_layers"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 81"
+  }
+}
+
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5_3"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 24   # 2(bg/fg) * 12(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 48   # 4 * 12(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16 \n'scales': !!python/tuple [4, 8, 16, 32]"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 24 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16 \n'scales': !!python/tuple [4, 8, 16, 32]"
+  }
+}
+
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 81"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 81
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 324
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
diff --git a/models/coco/VGG_CNN_M_1024/fast_rcnn/solver.prototxt b/models/coco/VGG_CNN_M_1024/fast_rcnn/solver.prototxt
new file mode 100644
index 0000000..af9aa44
--- /dev/null
+++ b/models/coco/VGG_CNN_M_1024/fast_rcnn/solver.prototxt
@@ -0,0 +1,15 @@
+train_net: "models/coco/VGG_CNN_M_1024/fast_rcnn/train.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 200000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg_cnn_m_1024_fast_rcnn"
+#debug_info: true
diff --git a/models/coco/VGG_CNN_M_1024/fast_rcnn/test.prototxt b/models/coco/VGG_CNN_M_1024/fast_rcnn/test.prototxt
new file mode 100644
index 0000000..733a759
--- /dev/null
+++ b/models/coco/VGG_CNN_M_1024/fast_rcnn/test.prototxt
@@ -0,0 +1,299 @@
+name: "VGG_CNN_M_1024"
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+input: "rois"
+input_shape {
+  dim: 1 # to be changed on-the-fly to num ROIs
+  dim: 5 # [batch ind, x1, y1, x2, y2] zero-based indexing
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7
+    stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 5
+    stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 81
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 324
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+}
diff --git a/models/coco/VGG_CNN_M_1024/fast_rcnn/train.prototxt b/models/coco/VGG_CNN_M_1024/fast_rcnn/train.prototxt
new file mode 100644
index 0000000..2f87439
--- /dev/null
+++ b/models/coco/VGG_CNN_M_1024/fast_rcnn/train.prototxt
@@ -0,0 +1,292 @@
+name: "VGG_CNN_M_1024"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 81"
+  }
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7
+    stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 5
+    stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 81
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 324
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
diff --git a/models/coco/VGG_CNN_M_1024/faster_rcnn_end2end/solver.prototxt b/models/coco/VGG_CNN_M_1024/faster_rcnn_end2end/solver.prototxt
new file mode 100644
index 0000000..c398a08
--- /dev/null
+++ b/models/coco/VGG_CNN_M_1024/faster_rcnn_end2end/solver.prototxt
@@ -0,0 +1,14 @@
+train_net: "models/coco/VGG_CNN_M_1024/faster_rcnn_end2end/train.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 350000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg_cnn_m_1024_faster_rcnn"
diff --git a/models/coco/VGG_CNN_M_1024/faster_rcnn_end2end/test.prototxt b/models/coco/VGG_CNN_M_1024/faster_rcnn_end2end/test.prototxt
new file mode 100644
index 0000000..ddc633e
--- /dev/null
+++ b/models/coco/VGG_CNN_M_1024/faster_rcnn_end2end/test.prototxt
@@ -0,0 +1,432 @@
+name: "VGG_CNN_M_1024"
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7
+    stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 5
+    stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+#layer {
+#  name: "rpn_conv/3x3"
+#  type: "Convolution"
+#  bottom: "conv5"
+#  top: "rpn_conv/3x3"
+#  param { lr_mult: 1.0 decay_mult: 1.0 }
+#  param { lr_mult: 2.0 decay_mult: 0 }
+#  convolution_param {
+#    num_output: 192
+#    kernel_size: 3 pad: 1 stride: 1
+#    weight_filler { type: "gaussian" std: 0.01 }
+#    bias_filler { type: "constant" value: 0 }
+#  }
+#}
+#layer {
+#  name: "rpn_conv/5x5"
+#  type: "Convolution"
+#  bottom: "conv5"
+#  top: "rpn_conv/5x5"
+#  param { lr_mult: 1.0 decay_mult: 1.0 }
+#  param { lr_mult: 2.0 decay_mult: 0 }
+#  convolution_param {
+#    num_output: 64
+#    kernel_size: 5 pad: 2 stride: 1
+#    weight_filler { type: "gaussian" std: 0.0036 }
+#    bias_filler { type: "constant" value: 0 }
+#  }
+#}
+#layer {
+#  name: "rpn/output"
+#  type: "Concat"
+#  bottom: "rpn_conv/3x3"
+#  bottom: "rpn_conv/5x5"
+#  top: "rpn/output"
+#}
+#layer {
+#  name: "rpn_relu/output"
+#  type: "ReLU"
+#  bottom: "rpn/output"
+#  top: "rpn/output"
+#}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 24  # 2(bg/fg) * 12(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 48   # 4 * 12(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 24 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16 \n'scales': !!python/tuple [4, 8, 16, 32]"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 81
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 324
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+}
diff --git a/models/coco/VGG_CNN_M_1024/faster_rcnn_end2end/train.prototxt b/models/coco/VGG_CNN_M_1024/faster_rcnn_end2end/train.prototxt
new file mode 100644
index 0000000..a15e7b3
--- /dev/null
+++ b/models/coco/VGG_CNN_M_1024/faster_rcnn_end2end/train.prototxt
@@ -0,0 +1,453 @@
+name: "VGG_CNN_M_1024"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 81"
+  }
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7
+    stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 5
+    stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 24   # 2(bg/fg) * 12(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 48   # 4 * 12(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16 \n'scales': !!python/tuple [4, 8, 16, 32]"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 24 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16 \n'scales': !!python/tuple [4, 8, 16, 32]"
+  }
+}
+
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 81"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 81
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 324
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
diff --git a/models/pascal_voc/ResNet-101/rfcn_end2end/class-aware/test.prototxt b/models/pascal_voc/ResNet-101/rfcn_end2end/class-aware/test.prototxt
new file mode 100644
index 0000000..076dd02
--- /dev/null
+++ b/models/pascal_voc/ResNet-101/rfcn_end2end/class-aware/test.prototxt
@@ -0,0 +1,7186 @@
+name: "ResNet101"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+# ------------------------ conv1 -----------------------------
+layer {
+    bottom: "data"
+    top: "conv1"
+    name: "conv1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 7
+        pad: 3
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "bn_conv1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "scale_conv1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "conv1"
+    bottom: "conv1"
+    name: "conv1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv1"
+    top: "pool1"
+    name: "pool1"
+    type: "Pooling"
+    pooling_param {
+        kernel_size: 3
+        stride: 2
+        pool: MAX
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch1"
+    name: "res2a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "bn2a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "scale2a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "bn2a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "scale2a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2a_branch2a"
+    bottom: "res2a_branch2a"
+    name: "res2a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "bn2a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "scale2a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2a_branch2b"
+    bottom: "res2a_branch2b"
+    name: "res2a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2c"
+    name: "res2a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "bn2a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "scale2a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    bottom: "res2a_branch2c"
+    top: "res2a"
+    name: "res2a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2a"
+    name: "res2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "bn2b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "scale2b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2b_branch2a"
+    bottom: "res2b_branch2a"
+    name: "res2b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "bn2b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "scale2b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2b_branch2b"
+    bottom: "res2b_branch2b"
+    name: "res2b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2c"
+    name: "res2b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "bn2b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "scale2b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a"
+    bottom: "res2b_branch2c"
+    top: "res2b"
+    name: "res2b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2b"
+    name: "res2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "bn2c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "scale2c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2c_branch2a"
+    bottom: "res2c_branch2a"
+    name: "res2c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "bn2c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "scale2c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2c_branch2b"
+    bottom: "res2c_branch2b"
+    name: "res2c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2c"
+    name: "res2c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "bn2c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "scale2c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b"
+    bottom: "res2c_branch2c"
+    top: "res2c"
+    name: "res2c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res2c"
+    name: "res2c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch1"
+    name: "res3a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "bn3a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "scale3a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "bn3a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "scale3a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3a_branch2a"
+    bottom: "res3a_branch2a"
+    name: "res3a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "bn3a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "scale3a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3a_branch2b"
+    bottom: "res3a_branch2b"
+    name: "res3a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2c"
+    name: "res3a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "bn3a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "scale3a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    bottom: "res3a_branch2c"
+    top: "res3a"
+    name: "res3a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3a"
+    name: "res3a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3b1_branch2a"
+    name: "res3b1_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2a"
+    name: "bn3b1_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2a"
+    name: "scale3b1_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b1_branch2a"
+    bottom: "res3b1_branch2a"
+    name: "res3b1_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2b"
+    name: "res3b1_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2b"
+    name: "bn3b1_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2b"
+    name: "scale3b1_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b1_branch2b"
+    bottom: "res3b1_branch2b"
+    name: "res3b1_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2c"
+    name: "res3b1_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2c"
+    top: "res3b1_branch2c"
+    name: "bn3b1_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2c"
+    top: "res3b1_branch2c"
+    name: "scale3b1_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a"
+    bottom: "res3b1_branch2c"
+    top: "res3b1"
+    name: "res3b1"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b1"
+    top: "res3b1"
+    name: "res3b1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1"
+    top: "res3b2_branch2a"
+    name: "res3b2_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2a"
+    name: "bn3b2_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2a"
+    name: "scale3b2_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b2_branch2a"
+    bottom: "res3b2_branch2a"
+    name: "res3b2_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2b"
+    name: "res3b2_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2b"
+    name: "bn3b2_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2b"
+    name: "scale3b2_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b2_branch2b"
+    bottom: "res3b2_branch2b"
+    name: "res3b2_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2c"
+    name: "res3b2_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2c"
+    top: "res3b2_branch2c"
+    name: "bn3b2_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2c"
+    top: "res3b2_branch2c"
+    name: "scale3b2_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1"
+    bottom: "res3b2_branch2c"
+    top: "res3b2"
+    name: "res3b2"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b2"
+    top: "res3b2"
+    name: "res3b2_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2"
+    top: "res3b3_branch2a"
+    name: "res3b3_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2a"
+    name: "bn3b3_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2a"
+    name: "scale3b3_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b3_branch2a"
+    bottom: "res3b3_branch2a"
+    name: "res3b3_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2b"
+    name: "res3b3_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2b"
+    name: "bn3b3_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2b"
+    name: "scale3b3_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b3_branch2b"
+    bottom: "res3b3_branch2b"
+    name: "res3b3_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2c"
+    name: "res3b3_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2c"
+    top: "res3b3_branch2c"
+    name: "bn3b3_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2c"
+    top: "res3b3_branch2c"
+    name: "scale3b3_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2"
+    bottom: "res3b3_branch2c"
+    top: "res3b3"
+    name: "res3b3"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res3b3"
+    name: "res3b3_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res4a_branch1"
+    name: "res4a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "bn4a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "scale4a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "bn4a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "scale4a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4a_branch2a"
+    bottom: "res4a_branch2a"
+    name: "res4a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "bn4a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "scale4a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4a_branch2b"
+    bottom: "res4a_branch2b"
+    name: "res4a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2c"
+    name: "res4a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "bn4a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "scale4a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    bottom: "res4a_branch2c"
+    top: "res4a"
+    name: "res4a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4a"
+    name: "res4a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4b1_branch2a"
+    name: "res4b1_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2a"
+    name: "bn4b1_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2a"
+    name: "scale4b1_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b1_branch2a"
+    bottom: "res4b1_branch2a"
+    name: "res4b1_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2b"
+    name: "res4b1_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2b"
+    name: "bn4b1_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2b"
+    name: "scale4b1_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b1_branch2b"
+    bottom: "res4b1_branch2b"
+    name: "res4b1_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2c"
+    name: "res4b1_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2c"
+    top: "res4b1_branch2c"
+    name: "bn4b1_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2c"
+    top: "res4b1_branch2c"
+    name: "scale4b1_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a"
+    bottom: "res4b1_branch2c"
+    top: "res4b1"
+    name: "res4b1"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b1"
+    top: "res4b1"
+    name: "res4b1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1"
+    top: "res4b2_branch2a"
+    name: "res4b2_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2a"
+    name: "bn4b2_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2a"
+    name: "scale4b2_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b2_branch2a"
+    bottom: "res4b2_branch2a"
+    name: "res4b2_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2b"
+    name: "res4b2_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2b"
+    name: "bn4b2_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2b"
+    name: "scale4b2_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b2_branch2b"
+    bottom: "res4b2_branch2b"
+    name: "res4b2_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2c"
+    name: "res4b2_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2c"
+    top: "res4b2_branch2c"
+    name: "bn4b2_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2c"
+    top: "res4b2_branch2c"
+    name: "scale4b2_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1"
+    bottom: "res4b2_branch2c"
+    top: "res4b2"
+    name: "res4b2"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b2"
+    top: "res4b2"
+    name: "res4b2_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2"
+    top: "res4b3_branch2a"
+    name: "res4b3_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2a"
+    name: "bn4b3_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2a"
+    name: "scale4b3_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b3_branch2a"
+    bottom: "res4b3_branch2a"
+    name: "res4b3_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2b"
+    name: "res4b3_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2b"
+    name: "bn4b3_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2b"
+    name: "scale4b3_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b3_branch2b"
+    bottom: "res4b3_branch2b"
+    name: "res4b3_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2c"
+    name: "res4b3_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2c"
+    top: "res4b3_branch2c"
+    name: "bn4b3_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2c"
+    top: "res4b3_branch2c"
+    name: "scale4b3_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2"
+    bottom: "res4b3_branch2c"
+    top: "res4b3"
+    name: "res4b3"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b3"
+    top: "res4b3"
+    name: "res4b3_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3"
+    top: "res4b4_branch2a"
+    name: "res4b4_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2a"
+    name: "bn4b4_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2a"
+    name: "scale4b4_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b4_branch2a"
+    bottom: "res4b4_branch2a"
+    name: "res4b4_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2b"
+    name: "res4b4_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2b"
+    name: "bn4b4_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2b"
+    name: "scale4b4_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b4_branch2b"
+    bottom: "res4b4_branch2b"
+    name: "res4b4_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2c"
+    name: "res4b4_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2c"
+    top: "res4b4_branch2c"
+    name: "bn4b4_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2c"
+    top: "res4b4_branch2c"
+    name: "scale4b4_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3"
+    bottom: "res4b4_branch2c"
+    top: "res4b4"
+    name: "res4b4"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b4"
+    top: "res4b4"
+    name: "res4b4_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4"
+    top: "res4b5_branch2a"
+    name: "res4b5_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2a"
+    name: "bn4b5_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2a"
+    name: "scale4b5_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b5_branch2a"
+    bottom: "res4b5_branch2a"
+    name: "res4b5_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2b"
+    name: "res4b5_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2b"
+    name: "bn4b5_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2b"
+    name: "scale4b5_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b5_branch2b"
+    bottom: "res4b5_branch2b"
+    name: "res4b5_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2c"
+    name: "res4b5_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2c"
+    top: "res4b5_branch2c"
+    name: "bn4b5_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2c"
+    top: "res4b5_branch2c"
+    name: "scale4b5_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4"
+    bottom: "res4b5_branch2c"
+    top: "res4b5"
+    name: "res4b5"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b5"
+    top: "res4b5"
+    name: "res4b5_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5"
+    top: "res4b6_branch2a"
+    name: "res4b6_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2a"
+    name: "bn4b6_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2a"
+    name: "scale4b6_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b6_branch2a"
+    bottom: "res4b6_branch2a"
+    name: "res4b6_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2b"
+    name: "res4b6_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2b"
+    name: "bn4b6_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2b"
+    name: "scale4b6_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b6_branch2b"
+    bottom: "res4b6_branch2b"
+    name: "res4b6_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2c"
+    name: "res4b6_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2c"
+    top: "res4b6_branch2c"
+    name: "bn4b6_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2c"
+    top: "res4b6_branch2c"
+    name: "scale4b6_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5"
+    bottom: "res4b6_branch2c"
+    top: "res4b6"
+    name: "res4b6"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b6"
+    top: "res4b6"
+    name: "res4b6_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6"
+    top: "res4b7_branch2a"
+    name: "res4b7_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2a"
+    name: "bn4b7_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2a"
+    name: "scale4b7_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b7_branch2a"
+    bottom: "res4b7_branch2a"
+    name: "res4b7_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2b"
+    name: "res4b7_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2b"
+    name: "bn4b7_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2b"
+    name: "scale4b7_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b7_branch2b"
+    bottom: "res4b7_branch2b"
+    name: "res4b7_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2c"
+    name: "res4b7_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2c"
+    top: "res4b7_branch2c"
+    name: "bn4b7_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2c"
+    top: "res4b7_branch2c"
+    name: "scale4b7_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6"
+    bottom: "res4b7_branch2c"
+    top: "res4b7"
+    name: "res4b7"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b7"
+    top: "res4b7"
+    name: "res4b7_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7"
+    top: "res4b8_branch2a"
+    name: "res4b8_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2a"
+    name: "bn4b8_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2a"
+    name: "scale4b8_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b8_branch2a"
+    bottom: "res4b8_branch2a"
+    name: "res4b8_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2b"
+    name: "res4b8_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2b"
+    name: "bn4b8_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2b"
+    name: "scale4b8_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b8_branch2b"
+    bottom: "res4b8_branch2b"
+    name: "res4b8_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2c"
+    name: "res4b8_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2c"
+    top: "res4b8_branch2c"
+    name: "bn4b8_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2c"
+    top: "res4b8_branch2c"
+    name: "scale4b8_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7"
+    bottom: "res4b8_branch2c"
+    top: "res4b8"
+    name: "res4b8"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b8"
+    top: "res4b8"
+    name: "res4b8_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8"
+    top: "res4b9_branch2a"
+    name: "res4b9_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2a"
+    name: "bn4b9_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2a"
+    name: "scale4b9_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b9_branch2a"
+    bottom: "res4b9_branch2a"
+    name: "res4b9_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2b"
+    name: "res4b9_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2b"
+    name: "bn4b9_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2b"
+    name: "scale4b9_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b9_branch2b"
+    bottom: "res4b9_branch2b"
+    name: "res4b9_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2c"
+    name: "res4b9_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2c"
+    top: "res4b9_branch2c"
+    name: "bn4b9_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2c"
+    top: "res4b9_branch2c"
+    name: "scale4b9_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8"
+    bottom: "res4b9_branch2c"
+    top: "res4b9"
+    name: "res4b9"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b9"
+    top: "res4b9"
+    name: "res4b9_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9"
+    top: "res4b10_branch2a"
+    name: "res4b10_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2a"
+    name: "bn4b10_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2a"
+    name: "scale4b10_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b10_branch2a"
+    bottom: "res4b10_branch2a"
+    name: "res4b10_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2b"
+    name: "res4b10_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2b"
+    name: "bn4b10_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2b"
+    name: "scale4b10_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b10_branch2b"
+    bottom: "res4b10_branch2b"
+    name: "res4b10_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2c"
+    name: "res4b10_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2c"
+    top: "res4b10_branch2c"
+    name: "bn4b10_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2c"
+    top: "res4b10_branch2c"
+    name: "scale4b10_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9"
+    bottom: "res4b10_branch2c"
+    top: "res4b10"
+    name: "res4b10"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b10"
+    top: "res4b10"
+    name: "res4b10_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10"
+    top: "res4b11_branch2a"
+    name: "res4b11_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2a"
+    name: "bn4b11_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2a"
+    name: "scale4b11_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b11_branch2a"
+    bottom: "res4b11_branch2a"
+    name: "res4b11_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2b"
+    name: "res4b11_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2b"
+    name: "bn4b11_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2b"
+    name: "scale4b11_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b11_branch2b"
+    bottom: "res4b11_branch2b"
+    name: "res4b11_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2c"
+    name: "res4b11_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2c"
+    top: "res4b11_branch2c"
+    name: "bn4b11_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2c"
+    top: "res4b11_branch2c"
+    name: "scale4b11_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10"
+    bottom: "res4b11_branch2c"
+    top: "res4b11"
+    name: "res4b11"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b11"
+    top: "res4b11"
+    name: "res4b11_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11"
+    top: "res4b12_branch2a"
+    name: "res4b12_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2a"
+    name: "bn4b12_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2a"
+    name: "scale4b12_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b12_branch2a"
+    bottom: "res4b12_branch2a"
+    name: "res4b12_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2b"
+    name: "res4b12_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2b"
+    name: "bn4b12_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2b"
+    name: "scale4b12_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b12_branch2b"
+    bottom: "res4b12_branch2b"
+    name: "res4b12_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2c"
+    name: "res4b12_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2c"
+    top: "res4b12_branch2c"
+    name: "bn4b12_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2c"
+    top: "res4b12_branch2c"
+    name: "scale4b12_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11"
+    bottom: "res4b12_branch2c"
+    top: "res4b12"
+    name: "res4b12"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b12"
+    top: "res4b12"
+    name: "res4b12_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12"
+    top: "res4b13_branch2a"
+    name: "res4b13_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2a"
+    name: "bn4b13_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2a"
+    name: "scale4b13_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b13_branch2a"
+    bottom: "res4b13_branch2a"
+    name: "res4b13_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2b"
+    name: "res4b13_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2b"
+    name: "bn4b13_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2b"
+    name: "scale4b13_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b13_branch2b"
+    bottom: "res4b13_branch2b"
+    name: "res4b13_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2c"
+    name: "res4b13_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2c"
+    top: "res4b13_branch2c"
+    name: "bn4b13_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2c"
+    top: "res4b13_branch2c"
+    name: "scale4b13_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12"
+    bottom: "res4b13_branch2c"
+    top: "res4b13"
+    name: "res4b13"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b13"
+    top: "res4b13"
+    name: "res4b13_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13"
+    top: "res4b14_branch2a"
+    name: "res4b14_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2a"
+    name: "bn4b14_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2a"
+    name: "scale4b14_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b14_branch2a"
+    bottom: "res4b14_branch2a"
+    name: "res4b14_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2b"
+    name: "res4b14_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2b"
+    name: "bn4b14_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2b"
+    name: "scale4b14_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b14_branch2b"
+    bottom: "res4b14_branch2b"
+    name: "res4b14_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2c"
+    name: "res4b14_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2c"
+    top: "res4b14_branch2c"
+    name: "bn4b14_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2c"
+    top: "res4b14_branch2c"
+    name: "scale4b14_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13"
+    bottom: "res4b14_branch2c"
+    top: "res4b14"
+    name: "res4b14"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b14"
+    top: "res4b14"
+    name: "res4b14_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14"
+    top: "res4b15_branch2a"
+    name: "res4b15_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2a"
+    name: "bn4b15_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2a"
+    name: "scale4b15_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b15_branch2a"
+    bottom: "res4b15_branch2a"
+    name: "res4b15_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2b"
+    name: "res4b15_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2b"
+    name: "bn4b15_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2b"
+    name: "scale4b15_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b15_branch2b"
+    bottom: "res4b15_branch2b"
+    name: "res4b15_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2c"
+    name: "res4b15_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2c"
+    top: "res4b15_branch2c"
+    name: "bn4b15_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2c"
+    top: "res4b15_branch2c"
+    name: "scale4b15_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14"
+    bottom: "res4b15_branch2c"
+    top: "res4b15"
+    name: "res4b15"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b15"
+    top: "res4b15"
+    name: "res4b15_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15"
+    top: "res4b16_branch2a"
+    name: "res4b16_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2a"
+    name: "bn4b16_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2a"
+    name: "scale4b16_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b16_branch2a"
+    bottom: "res4b16_branch2a"
+    name: "res4b16_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2b"
+    name: "res4b16_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2b"
+    name: "bn4b16_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2b"
+    name: "scale4b16_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b16_branch2b"
+    bottom: "res4b16_branch2b"
+    name: "res4b16_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2c"
+    name: "res4b16_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2c"
+    top: "res4b16_branch2c"
+    name: "bn4b16_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2c"
+    top: "res4b16_branch2c"
+    name: "scale4b16_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15"
+    bottom: "res4b16_branch2c"
+    top: "res4b16"
+    name: "res4b16"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b16"
+    top: "res4b16"
+    name: "res4b16_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16"
+    top: "res4b17_branch2a"
+    name: "res4b17_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2a"
+    name: "bn4b17_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2a"
+    name: "scale4b17_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b17_branch2a"
+    bottom: "res4b17_branch2a"
+    name: "res4b17_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2b"
+    name: "res4b17_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2b"
+    name: "bn4b17_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2b"
+    name: "scale4b17_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b17_branch2b"
+    bottom: "res4b17_branch2b"
+    name: "res4b17_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2c"
+    name: "res4b17_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2c"
+    top: "res4b17_branch2c"
+    name: "bn4b17_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2c"
+    top: "res4b17_branch2c"
+    name: "scale4b17_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16"
+    bottom: "res4b17_branch2c"
+    top: "res4b17"
+    name: "res4b17"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b17"
+    top: "res4b17"
+    name: "res4b17_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17"
+    top: "res4b18_branch2a"
+    name: "res4b18_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2a"
+    name: "bn4b18_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2a"
+    name: "scale4b18_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b18_branch2a"
+    bottom: "res4b18_branch2a"
+    name: "res4b18_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2b"
+    name: "res4b18_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2b"
+    name: "bn4b18_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2b"
+    name: "scale4b18_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b18_branch2b"
+    bottom: "res4b18_branch2b"
+    name: "res4b18_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2c"
+    name: "res4b18_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2c"
+    top: "res4b18_branch2c"
+    name: "bn4b18_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2c"
+    top: "res4b18_branch2c"
+    name: "scale4b18_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17"
+    bottom: "res4b18_branch2c"
+    top: "res4b18"
+    name: "res4b18"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b18"
+    top: "res4b18"
+    name: "res4b18_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18"
+    top: "res4b19_branch2a"
+    name: "res4b19_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2a"
+    name: "bn4b19_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2a"
+    name: "scale4b19_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b19_branch2a"
+    bottom: "res4b19_branch2a"
+    name: "res4b19_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2b"
+    name: "res4b19_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2b"
+    name: "bn4b19_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2b"
+    name: "scale4b19_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b19_branch2b"
+    bottom: "res4b19_branch2b"
+    name: "res4b19_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2c"
+    name: "res4b19_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2c"
+    top: "res4b19_branch2c"
+    name: "bn4b19_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2c"
+    top: "res4b19_branch2c"
+    name: "scale4b19_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18"
+    bottom: "res4b19_branch2c"
+    top: "res4b19"
+    name: "res4b19"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b19"
+    top: "res4b19"
+    name: "res4b19_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19"
+    top: "res4b20_branch2a"
+    name: "res4b20_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2a"
+    name: "bn4b20_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2a"
+    name: "scale4b20_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b20_branch2a"
+    bottom: "res4b20_branch2a"
+    name: "res4b20_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2b"
+    name: "res4b20_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2b"
+    name: "bn4b20_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2b"
+    name: "scale4b20_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b20_branch2b"
+    bottom: "res4b20_branch2b"
+    name: "res4b20_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2c"
+    name: "res4b20_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2c"
+    top: "res4b20_branch2c"
+    name: "bn4b20_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2c"
+    top: "res4b20_branch2c"
+    name: "scale4b20_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19"
+    bottom: "res4b20_branch2c"
+    top: "res4b20"
+    name: "res4b20"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b20"
+    top: "res4b20"
+    name: "res4b20_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20"
+    top: "res4b21_branch2a"
+    name: "res4b21_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2a"
+    name: "bn4b21_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2a"
+    name: "scale4b21_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b21_branch2a"
+    bottom: "res4b21_branch2a"
+    name: "res4b21_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2b"
+    name: "res4b21_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2b"
+    name: "bn4b21_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2b"
+    name: "scale4b21_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b21_branch2b"
+    bottom: "res4b21_branch2b"
+    name: "res4b21_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2c"
+    name: "res4b21_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2c"
+    top: "res4b21_branch2c"
+    name: "bn4b21_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2c"
+    top: "res4b21_branch2c"
+    name: "scale4b21_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20"
+    bottom: "res4b21_branch2c"
+    top: "res4b21"
+    name: "res4b21"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b21"
+    top: "res4b21"
+    name: "res4b21_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21"
+    top: "res4b22_branch2a"
+    name: "res4b22_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2a"
+    name: "bn4b22_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2a"
+    name: "scale4b22_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b22_branch2a"
+    bottom: "res4b22_branch2a"
+    name: "res4b22_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2b"
+    name: "res4b22_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2b"
+    name: "bn4b22_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2b"
+    name: "scale4b22_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b22_branch2b"
+    bottom: "res4b22_branch2b"
+    name: "res4b22_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2c"
+    name: "res4b22_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2c"
+    top: "res4b22_branch2c"
+    name: "bn4b22_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2c"
+    top: "res4b22_branch2c"
+    name: "scale4b22_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21"
+    bottom: "res4b22_branch2c"
+    top: "res4b22"
+    name: "res4b22"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res4b22"
+    name: "res4b22_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res5a_branch1"
+    name: "res5a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "bn5a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "scale5a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "bn5a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "scale5a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5a_branch2a"
+    bottom: "res5a_branch2a"
+    name: "res5a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "bn5a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "scale5a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5a_branch2b"
+    bottom: "res5a_branch2b"
+    name: "res5a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2c"
+    name: "res5a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "bn5a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "scale5a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    bottom: "res5a_branch2c"
+    top: "res5a"
+    name: "res5a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5a"
+    name: "res5a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "bn5b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "scale5b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5b_branch2a"
+    bottom: "res5b_branch2a"
+    name: "res5b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "bn5b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "scale5b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5b_branch2b"
+    bottom: "res5b_branch2b"
+    name: "res5b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2c"
+    name: "res5b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "bn5b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "scale5b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a"
+    bottom: "res5b_branch2c"
+    top: "res5b"
+    name: "res5b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5b"
+    name: "res5b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "bn5c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "scale5c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5c_branch2a"
+    bottom: "res5c_branch2a"
+    name: "res5c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "bn5c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "scale5c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5c_branch2b"
+    bottom: "res5c_branch2b"
+    name: "res5c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2c"
+    name: "res5c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "bn5c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "scale5c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b"
+    bottom: "res5c_branch2c"
+    top: "res5c"
+    name: "res5c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5c"
+    top: "res5c"
+    name: "res5c_relu"
+    type: "ReLU"
+}
+
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "res5c"
+  top: "rpn/output"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#----------------------new conv layer------------------
+layer {
+    bottom: "res5c"
+    top: "conv_new_1"
+    name: "conv_new_1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "conv_new_1"
+    name: "conv_new_1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_cls"
+    name: "rfcn_cls"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1029 #21*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_bbox"
+    name: "rfcn_bbox"
+    type: "Convolution"
+    convolution_param {
+        num_output: 4116 #8*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+#--------------position sensitive RoI pooling--------------
+layer {
+    bottom: "rfcn_cls"
+    bottom: "rois"
+    top: "psroipooled_cls_rois"
+    name: "psroipooled_cls_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 21
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_cls_rois"
+    top: "cls_score"
+    name: "ave_cls_score_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+layer {
+    bottom: "rfcn_bbox"
+    bottom: "rois"
+    top: "psroipooled_loc_rois"
+    name: "psroipooled_loc_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 84
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_loc_rois"
+    top: "bbox_pred_pre"
+    name: "ave_bbox_pred_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+#-----------------------output------------------------
+layer {
+    name: "cls_prob"
+    type: "Softmax"
+    bottom: "cls_score"
+    top: "cls_prob_pre"
+}
+
+layer {
+    name: "cls_prob_reshape"
+    type: "Reshape"
+    bottom: "cls_prob_pre"
+    top: "cls_prob"
+    reshape_param {
+        shape {
+            dim: -1
+            dim: 21
+        }
+    }
+}
+
+layer {
+    name: "bbox_pred_reshape"
+    type: "Reshape"
+    bottom: "bbox_pred_pre"
+    top: "bbox_pred"
+    reshape_param {
+        shape {
+            dim: -1
+            dim: 84
+        }
+    }
+}
+
+
diff --git a/models/pascal_voc/ResNet-101/rfcn_end2end/class-aware/train_ohem.prototxt b/models/pascal_voc/ResNet-101/rfcn_end2end/class-aware/train_ohem.prototxt
new file mode 100644
index 0000000..ae3ce44
--- /dev/null
+++ b/models/pascal_voc/ResNet-101/rfcn_end2end/class-aware/train_ohem.prototxt
@@ -0,0 +1,7343 @@
+name: "ResNet-101"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+# ------------------------ conv1 -----------------------------
+layer {
+    bottom: "data"
+    top: "conv1"
+    name: "conv1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 7
+        pad: 3
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "bn_conv1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "scale_conv1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "conv1"
+    bottom: "conv1"
+    name: "conv1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv1"
+    top: "pool1"
+    name: "pool1"
+    type: "Pooling"
+    pooling_param {
+        kernel_size: 3
+        stride: 2
+        pool: MAX
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch1"
+    name: "res2a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "bn2a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "scale2a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "bn2a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "scale2a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2a_branch2a"
+    bottom: "res2a_branch2a"
+    name: "res2a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "bn2a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "scale2a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2a_branch2b"
+    bottom: "res2a_branch2b"
+    name: "res2a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2c"
+    name: "res2a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "bn2a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "scale2a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    bottom: "res2a_branch2c"
+    top: "res2a"
+    name: "res2a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2a"
+    name: "res2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "bn2b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "scale2b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2b_branch2a"
+    bottom: "res2b_branch2a"
+    name: "res2b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "bn2b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "scale2b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2b_branch2b"
+    bottom: "res2b_branch2b"
+    name: "res2b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2c"
+    name: "res2b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "bn2b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "scale2b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a"
+    bottom: "res2b_branch2c"
+    top: "res2b"
+    name: "res2b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2b"
+    name: "res2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "bn2c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "scale2c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2c_branch2a"
+    bottom: "res2c_branch2a"
+    name: "res2c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "bn2c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "scale2c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2c_branch2b"
+    bottom: "res2c_branch2b"
+    name: "res2c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2c"
+    name: "res2c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "bn2c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "scale2c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b"
+    bottom: "res2c_branch2c"
+    top: "res2c"
+    name: "res2c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res2c"
+    name: "res2c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch1"
+    name: "res3a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "bn3a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "scale3a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "bn3a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "scale3a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3a_branch2a"
+    bottom: "res3a_branch2a"
+    name: "res3a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "bn3a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "scale3a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3a_branch2b"
+    bottom: "res3a_branch2b"
+    name: "res3a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2c"
+    name: "res3a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "bn3a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "scale3a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    bottom: "res3a_branch2c"
+    top: "res3a"
+    name: "res3a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3a"
+    name: "res3a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3b1_branch2a"
+    name: "res3b1_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2a"
+    name: "bn3b1_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2a"
+    name: "scale3b1_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b1_branch2a"
+    bottom: "res3b1_branch2a"
+    name: "res3b1_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2b"
+    name: "res3b1_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2b"
+    name: "bn3b1_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2b"
+    name: "scale3b1_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b1_branch2b"
+    bottom: "res3b1_branch2b"
+    name: "res3b1_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2c"
+    name: "res3b1_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2c"
+    top: "res3b1_branch2c"
+    name: "bn3b1_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2c"
+    top: "res3b1_branch2c"
+    name: "scale3b1_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a"
+    bottom: "res3b1_branch2c"
+    top: "res3b1"
+    name: "res3b1"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b1"
+    top: "res3b1"
+    name: "res3b1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1"
+    top: "res3b2_branch2a"
+    name: "res3b2_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2a"
+    name: "bn3b2_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2a"
+    name: "scale3b2_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b2_branch2a"
+    bottom: "res3b2_branch2a"
+    name: "res3b2_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2b"
+    name: "res3b2_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2b"
+    name: "bn3b2_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2b"
+    name: "scale3b2_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b2_branch2b"
+    bottom: "res3b2_branch2b"
+    name: "res3b2_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2c"
+    name: "res3b2_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2c"
+    top: "res3b2_branch2c"
+    name: "bn3b2_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2c"
+    top: "res3b2_branch2c"
+    name: "scale3b2_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1"
+    bottom: "res3b2_branch2c"
+    top: "res3b2"
+    name: "res3b2"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b2"
+    top: "res3b2"
+    name: "res3b2_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2"
+    top: "res3b3_branch2a"
+    name: "res3b3_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2a"
+    name: "bn3b3_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2a"
+    name: "scale3b3_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b3_branch2a"
+    bottom: "res3b3_branch2a"
+    name: "res3b3_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2b"
+    name: "res3b3_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2b"
+    name: "bn3b3_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2b"
+    name: "scale3b3_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b3_branch2b"
+    bottom: "res3b3_branch2b"
+    name: "res3b3_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2c"
+    name: "res3b3_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2c"
+    top: "res3b3_branch2c"
+    name: "bn3b3_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2c"
+    top: "res3b3_branch2c"
+    name: "scale3b3_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2"
+    bottom: "res3b3_branch2c"
+    top: "res3b3"
+    name: "res3b3"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res3b3"
+    name: "res3b3_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res4a_branch1"
+    name: "res4a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "bn4a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "scale4a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "bn4a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "scale4a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4a_branch2a"
+    bottom: "res4a_branch2a"
+    name: "res4a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "bn4a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "scale4a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4a_branch2b"
+    bottom: "res4a_branch2b"
+    name: "res4a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2c"
+    name: "res4a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "bn4a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "scale4a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    bottom: "res4a_branch2c"
+    top: "res4a"
+    name: "res4a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4a"
+    name: "res4a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4b1_branch2a"
+    name: "res4b1_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2a"
+    name: "bn4b1_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2a"
+    name: "scale4b1_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b1_branch2a"
+    bottom: "res4b1_branch2a"
+    name: "res4b1_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2b"
+    name: "res4b1_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2b"
+    name: "bn4b1_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2b"
+    name: "scale4b1_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b1_branch2b"
+    bottom: "res4b1_branch2b"
+    name: "res4b1_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2c"
+    name: "res4b1_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2c"
+    top: "res4b1_branch2c"
+    name: "bn4b1_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2c"
+    top: "res4b1_branch2c"
+    name: "scale4b1_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a"
+    bottom: "res4b1_branch2c"
+    top: "res4b1"
+    name: "res4b1"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b1"
+    top: "res4b1"
+    name: "res4b1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1"
+    top: "res4b2_branch2a"
+    name: "res4b2_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2a"
+    name: "bn4b2_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2a"
+    name: "scale4b2_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b2_branch2a"
+    bottom: "res4b2_branch2a"
+    name: "res4b2_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2b"
+    name: "res4b2_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2b"
+    name: "bn4b2_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2b"
+    name: "scale4b2_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b2_branch2b"
+    bottom: "res4b2_branch2b"
+    name: "res4b2_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2c"
+    name: "res4b2_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2c"
+    top: "res4b2_branch2c"
+    name: "bn4b2_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2c"
+    top: "res4b2_branch2c"
+    name: "scale4b2_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1"
+    bottom: "res4b2_branch2c"
+    top: "res4b2"
+    name: "res4b2"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b2"
+    top: "res4b2"
+    name: "res4b2_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2"
+    top: "res4b3_branch2a"
+    name: "res4b3_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2a"
+    name: "bn4b3_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2a"
+    name: "scale4b3_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b3_branch2a"
+    bottom: "res4b3_branch2a"
+    name: "res4b3_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2b"
+    name: "res4b3_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2b"
+    name: "bn4b3_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2b"
+    name: "scale4b3_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b3_branch2b"
+    bottom: "res4b3_branch2b"
+    name: "res4b3_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2c"
+    name: "res4b3_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2c"
+    top: "res4b3_branch2c"
+    name: "bn4b3_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2c"
+    top: "res4b3_branch2c"
+    name: "scale4b3_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2"
+    bottom: "res4b3_branch2c"
+    top: "res4b3"
+    name: "res4b3"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b3"
+    top: "res4b3"
+    name: "res4b3_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3"
+    top: "res4b4_branch2a"
+    name: "res4b4_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2a"
+    name: "bn4b4_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2a"
+    name: "scale4b4_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b4_branch2a"
+    bottom: "res4b4_branch2a"
+    name: "res4b4_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2b"
+    name: "res4b4_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2b"
+    name: "bn4b4_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2b"
+    name: "scale4b4_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b4_branch2b"
+    bottom: "res4b4_branch2b"
+    name: "res4b4_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2c"
+    name: "res4b4_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2c"
+    top: "res4b4_branch2c"
+    name: "bn4b4_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2c"
+    top: "res4b4_branch2c"
+    name: "scale4b4_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3"
+    bottom: "res4b4_branch2c"
+    top: "res4b4"
+    name: "res4b4"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b4"
+    top: "res4b4"
+    name: "res4b4_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4"
+    top: "res4b5_branch2a"
+    name: "res4b5_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2a"
+    name: "bn4b5_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2a"
+    name: "scale4b5_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b5_branch2a"
+    bottom: "res4b5_branch2a"
+    name: "res4b5_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2b"
+    name: "res4b5_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2b"
+    name: "bn4b5_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2b"
+    name: "scale4b5_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b5_branch2b"
+    bottom: "res4b5_branch2b"
+    name: "res4b5_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2c"
+    name: "res4b5_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2c"
+    top: "res4b5_branch2c"
+    name: "bn4b5_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2c"
+    top: "res4b5_branch2c"
+    name: "scale4b5_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4"
+    bottom: "res4b5_branch2c"
+    top: "res4b5"
+    name: "res4b5"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b5"
+    top: "res4b5"
+    name: "res4b5_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5"
+    top: "res4b6_branch2a"
+    name: "res4b6_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2a"
+    name: "bn4b6_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2a"
+    name: "scale4b6_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b6_branch2a"
+    bottom: "res4b6_branch2a"
+    name: "res4b6_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2b"
+    name: "res4b6_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2b"
+    name: "bn4b6_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2b"
+    name: "scale4b6_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b6_branch2b"
+    bottom: "res4b6_branch2b"
+    name: "res4b6_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2c"
+    name: "res4b6_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2c"
+    top: "res4b6_branch2c"
+    name: "bn4b6_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2c"
+    top: "res4b6_branch2c"
+    name: "scale4b6_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5"
+    bottom: "res4b6_branch2c"
+    top: "res4b6"
+    name: "res4b6"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b6"
+    top: "res4b6"
+    name: "res4b6_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6"
+    top: "res4b7_branch2a"
+    name: "res4b7_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2a"
+    name: "bn4b7_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2a"
+    name: "scale4b7_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b7_branch2a"
+    bottom: "res4b7_branch2a"
+    name: "res4b7_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2b"
+    name: "res4b7_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2b"
+    name: "bn4b7_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2b"
+    name: "scale4b7_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b7_branch2b"
+    bottom: "res4b7_branch2b"
+    name: "res4b7_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2c"
+    name: "res4b7_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2c"
+    top: "res4b7_branch2c"
+    name: "bn4b7_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2c"
+    top: "res4b7_branch2c"
+    name: "scale4b7_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6"
+    bottom: "res4b7_branch2c"
+    top: "res4b7"
+    name: "res4b7"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b7"
+    top: "res4b7"
+    name: "res4b7_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7"
+    top: "res4b8_branch2a"
+    name: "res4b8_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2a"
+    name: "bn4b8_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2a"
+    name: "scale4b8_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b8_branch2a"
+    bottom: "res4b8_branch2a"
+    name: "res4b8_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2b"
+    name: "res4b8_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2b"
+    name: "bn4b8_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2b"
+    name: "scale4b8_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b8_branch2b"
+    bottom: "res4b8_branch2b"
+    name: "res4b8_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2c"
+    name: "res4b8_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2c"
+    top: "res4b8_branch2c"
+    name: "bn4b8_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2c"
+    top: "res4b8_branch2c"
+    name: "scale4b8_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7"
+    bottom: "res4b8_branch2c"
+    top: "res4b8"
+    name: "res4b8"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b8"
+    top: "res4b8"
+    name: "res4b8_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8"
+    top: "res4b9_branch2a"
+    name: "res4b9_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2a"
+    name: "bn4b9_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2a"
+    name: "scale4b9_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b9_branch2a"
+    bottom: "res4b9_branch2a"
+    name: "res4b9_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2b"
+    name: "res4b9_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2b"
+    name: "bn4b9_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2b"
+    name: "scale4b9_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b9_branch2b"
+    bottom: "res4b9_branch2b"
+    name: "res4b9_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2c"
+    name: "res4b9_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2c"
+    top: "res4b9_branch2c"
+    name: "bn4b9_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2c"
+    top: "res4b9_branch2c"
+    name: "scale4b9_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8"
+    bottom: "res4b9_branch2c"
+    top: "res4b9"
+    name: "res4b9"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b9"
+    top: "res4b9"
+    name: "res4b9_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9"
+    top: "res4b10_branch2a"
+    name: "res4b10_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2a"
+    name: "bn4b10_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2a"
+    name: "scale4b10_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b10_branch2a"
+    bottom: "res4b10_branch2a"
+    name: "res4b10_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2b"
+    name: "res4b10_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2b"
+    name: "bn4b10_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2b"
+    name: "scale4b10_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b10_branch2b"
+    bottom: "res4b10_branch2b"
+    name: "res4b10_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2c"
+    name: "res4b10_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2c"
+    top: "res4b10_branch2c"
+    name: "bn4b10_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2c"
+    top: "res4b10_branch2c"
+    name: "scale4b10_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9"
+    bottom: "res4b10_branch2c"
+    top: "res4b10"
+    name: "res4b10"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b10"
+    top: "res4b10"
+    name: "res4b10_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10"
+    top: "res4b11_branch2a"
+    name: "res4b11_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2a"
+    name: "bn4b11_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2a"
+    name: "scale4b11_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b11_branch2a"
+    bottom: "res4b11_branch2a"
+    name: "res4b11_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2b"
+    name: "res4b11_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2b"
+    name: "bn4b11_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2b"
+    name: "scale4b11_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b11_branch2b"
+    bottom: "res4b11_branch2b"
+    name: "res4b11_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2c"
+    name: "res4b11_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2c"
+    top: "res4b11_branch2c"
+    name: "bn4b11_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2c"
+    top: "res4b11_branch2c"
+    name: "scale4b11_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10"
+    bottom: "res4b11_branch2c"
+    top: "res4b11"
+    name: "res4b11"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b11"
+    top: "res4b11"
+    name: "res4b11_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11"
+    top: "res4b12_branch2a"
+    name: "res4b12_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2a"
+    name: "bn4b12_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2a"
+    name: "scale4b12_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b12_branch2a"
+    bottom: "res4b12_branch2a"
+    name: "res4b12_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2b"
+    name: "res4b12_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2b"
+    name: "bn4b12_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2b"
+    name: "scale4b12_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b12_branch2b"
+    bottom: "res4b12_branch2b"
+    name: "res4b12_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2c"
+    name: "res4b12_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2c"
+    top: "res4b12_branch2c"
+    name: "bn4b12_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2c"
+    top: "res4b12_branch2c"
+    name: "scale4b12_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11"
+    bottom: "res4b12_branch2c"
+    top: "res4b12"
+    name: "res4b12"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b12"
+    top: "res4b12"
+    name: "res4b12_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12"
+    top: "res4b13_branch2a"
+    name: "res4b13_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2a"
+    name: "bn4b13_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2a"
+    name: "scale4b13_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b13_branch2a"
+    bottom: "res4b13_branch2a"
+    name: "res4b13_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2b"
+    name: "res4b13_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2b"
+    name: "bn4b13_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2b"
+    name: "scale4b13_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b13_branch2b"
+    bottom: "res4b13_branch2b"
+    name: "res4b13_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2c"
+    name: "res4b13_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2c"
+    top: "res4b13_branch2c"
+    name: "bn4b13_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2c"
+    top: "res4b13_branch2c"
+    name: "scale4b13_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12"
+    bottom: "res4b13_branch2c"
+    top: "res4b13"
+    name: "res4b13"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b13"
+    top: "res4b13"
+    name: "res4b13_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13"
+    top: "res4b14_branch2a"
+    name: "res4b14_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2a"
+    name: "bn4b14_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2a"
+    name: "scale4b14_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b14_branch2a"
+    bottom: "res4b14_branch2a"
+    name: "res4b14_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2b"
+    name: "res4b14_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2b"
+    name: "bn4b14_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2b"
+    name: "scale4b14_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b14_branch2b"
+    bottom: "res4b14_branch2b"
+    name: "res4b14_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2c"
+    name: "res4b14_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2c"
+    top: "res4b14_branch2c"
+    name: "bn4b14_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2c"
+    top: "res4b14_branch2c"
+    name: "scale4b14_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13"
+    bottom: "res4b14_branch2c"
+    top: "res4b14"
+    name: "res4b14"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b14"
+    top: "res4b14"
+    name: "res4b14_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14"
+    top: "res4b15_branch2a"
+    name: "res4b15_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2a"
+    name: "bn4b15_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2a"
+    name: "scale4b15_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b15_branch2a"
+    bottom: "res4b15_branch2a"
+    name: "res4b15_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2b"
+    name: "res4b15_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2b"
+    name: "bn4b15_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2b"
+    name: "scale4b15_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b15_branch2b"
+    bottom: "res4b15_branch2b"
+    name: "res4b15_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2c"
+    name: "res4b15_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2c"
+    top: "res4b15_branch2c"
+    name: "bn4b15_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2c"
+    top: "res4b15_branch2c"
+    name: "scale4b15_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14"
+    bottom: "res4b15_branch2c"
+    top: "res4b15"
+    name: "res4b15"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b15"
+    top: "res4b15"
+    name: "res4b15_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15"
+    top: "res4b16_branch2a"
+    name: "res4b16_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2a"
+    name: "bn4b16_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2a"
+    name: "scale4b16_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b16_branch2a"
+    bottom: "res4b16_branch2a"
+    name: "res4b16_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2b"
+    name: "res4b16_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2b"
+    name: "bn4b16_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2b"
+    name: "scale4b16_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b16_branch2b"
+    bottom: "res4b16_branch2b"
+    name: "res4b16_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2c"
+    name: "res4b16_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2c"
+    top: "res4b16_branch2c"
+    name: "bn4b16_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2c"
+    top: "res4b16_branch2c"
+    name: "scale4b16_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15"
+    bottom: "res4b16_branch2c"
+    top: "res4b16"
+    name: "res4b16"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b16"
+    top: "res4b16"
+    name: "res4b16_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16"
+    top: "res4b17_branch2a"
+    name: "res4b17_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2a"
+    name: "bn4b17_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2a"
+    name: "scale4b17_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b17_branch2a"
+    bottom: "res4b17_branch2a"
+    name: "res4b17_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2b"
+    name: "res4b17_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2b"
+    name: "bn4b17_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2b"
+    name: "scale4b17_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b17_branch2b"
+    bottom: "res4b17_branch2b"
+    name: "res4b17_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2c"
+    name: "res4b17_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2c"
+    top: "res4b17_branch2c"
+    name: "bn4b17_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2c"
+    top: "res4b17_branch2c"
+    name: "scale4b17_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16"
+    bottom: "res4b17_branch2c"
+    top: "res4b17"
+    name: "res4b17"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b17"
+    top: "res4b17"
+    name: "res4b17_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17"
+    top: "res4b18_branch2a"
+    name: "res4b18_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2a"
+    name: "bn4b18_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2a"
+    name: "scale4b18_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b18_branch2a"
+    bottom: "res4b18_branch2a"
+    name: "res4b18_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2b"
+    name: "res4b18_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2b"
+    name: "bn4b18_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2b"
+    name: "scale4b18_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b18_branch2b"
+    bottom: "res4b18_branch2b"
+    name: "res4b18_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2c"
+    name: "res4b18_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2c"
+    top: "res4b18_branch2c"
+    name: "bn4b18_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2c"
+    top: "res4b18_branch2c"
+    name: "scale4b18_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17"
+    bottom: "res4b18_branch2c"
+    top: "res4b18"
+    name: "res4b18"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b18"
+    top: "res4b18"
+    name: "res4b18_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18"
+    top: "res4b19_branch2a"
+    name: "res4b19_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2a"
+    name: "bn4b19_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2a"
+    name: "scale4b19_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b19_branch2a"
+    bottom: "res4b19_branch2a"
+    name: "res4b19_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2b"
+    name: "res4b19_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2b"
+    name: "bn4b19_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2b"
+    name: "scale4b19_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b19_branch2b"
+    bottom: "res4b19_branch2b"
+    name: "res4b19_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2c"
+    name: "res4b19_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2c"
+    top: "res4b19_branch2c"
+    name: "bn4b19_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2c"
+    top: "res4b19_branch2c"
+    name: "scale4b19_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18"
+    bottom: "res4b19_branch2c"
+    top: "res4b19"
+    name: "res4b19"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b19"
+    top: "res4b19"
+    name: "res4b19_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19"
+    top: "res4b20_branch2a"
+    name: "res4b20_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2a"
+    name: "bn4b20_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2a"
+    name: "scale4b20_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b20_branch2a"
+    bottom: "res4b20_branch2a"
+    name: "res4b20_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2b"
+    name: "res4b20_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2b"
+    name: "bn4b20_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2b"
+    name: "scale4b20_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b20_branch2b"
+    bottom: "res4b20_branch2b"
+    name: "res4b20_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2c"
+    name: "res4b20_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2c"
+    top: "res4b20_branch2c"
+    name: "bn4b20_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2c"
+    top: "res4b20_branch2c"
+    name: "scale4b20_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19"
+    bottom: "res4b20_branch2c"
+    top: "res4b20"
+    name: "res4b20"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b20"
+    top: "res4b20"
+    name: "res4b20_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20"
+    top: "res4b21_branch2a"
+    name: "res4b21_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2a"
+    name: "bn4b21_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2a"
+    name: "scale4b21_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b21_branch2a"
+    bottom: "res4b21_branch2a"
+    name: "res4b21_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2b"
+    name: "res4b21_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2b"
+    name: "bn4b21_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2b"
+    name: "scale4b21_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b21_branch2b"
+    bottom: "res4b21_branch2b"
+    name: "res4b21_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2c"
+    name: "res4b21_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2c"
+    top: "res4b21_branch2c"
+    name: "bn4b21_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2c"
+    top: "res4b21_branch2c"
+    name: "scale4b21_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20"
+    bottom: "res4b21_branch2c"
+    top: "res4b21"
+    name: "res4b21"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b21"
+    top: "res4b21"
+    name: "res4b21_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21"
+    top: "res4b22_branch2a"
+    name: "res4b22_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2a"
+    name: "bn4b22_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2a"
+    name: "scale4b22_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b22_branch2a"
+    bottom: "res4b22_branch2a"
+    name: "res4b22_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2b"
+    name: "res4b22_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2b"
+    name: "bn4b22_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2b"
+    name: "scale4b22_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b22_branch2b"
+    bottom: "res4b22_branch2b"
+    name: "res4b22_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2c"
+    name: "res4b22_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2c"
+    top: "res4b22_branch2c"
+    name: "bn4b22_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2c"
+    top: "res4b22_branch2c"
+    name: "scale4b22_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21"
+    bottom: "res4b22_branch2c"
+    top: "res4b22"
+    name: "res4b22"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res4b22"
+    name: "res4b22_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res5a_branch1"
+    name: "res5a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "bn5a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "scale5a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "bn5a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "scale5a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5a_branch2a"
+    bottom: "res5a_branch2a"
+    name: "res5a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "bn5a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "scale5a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5a_branch2b"
+    bottom: "res5a_branch2b"
+    name: "res5a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2c"
+    name: "res5a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "bn5a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "scale5a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    bottom: "res5a_branch2c"
+    top: "res5a"
+    name: "res5a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5a"
+    name: "res5a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "bn5b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "scale5b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5b_branch2a"
+    bottom: "res5b_branch2a"
+    name: "res5b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "bn5b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "scale5b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5b_branch2b"
+    bottom: "res5b_branch2b"
+    name: "res5b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2c"
+    name: "res5b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "bn5b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "scale5b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a"
+    bottom: "res5b_branch2c"
+    top: "res5b"
+    name: "res5b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5b"
+    name: "res5b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "bn5c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "scale5c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5c_branch2a"
+    bottom: "res5c_branch2a"
+    name: "res5c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "bn5c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "scale5c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5c_branch2b"
+    bottom: "res5c_branch2b"
+    name: "res5c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2c"
+    name: "res5c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "bn5c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "scale5c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b"
+    bottom: "res5c_branch2c"
+    top: "res5c"
+    name: "res5c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5c"
+    top: "res5c"
+    name: "res5c_relu"
+    type: "ReLU"
+}
+
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "res5c"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+#  top: 'rpn_scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#layer {
+#  name: 'debug-data'
+#  type: 'Python'
+#  bottom: 'data'
+#  bottom: 'rpn_rois'
+#  bottom: 'rpn_scores'
+#  python_param {
+#    module: 'rpn.debug_layer'
+#    layer: 'RPNDebugLayer'
+#  }
+#}
+
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#----------------------new conv layer------------------
+layer {
+    bottom: "res5c"
+    top: "conv_new_1"
+    name: "conv_new_1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "conv_new_1"
+    name: "conv_new_1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_cls"
+    name: "rfcn_cls"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1029 #21*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_bbox"
+    name: "rfcn_bbox"
+    type: "Convolution"
+    convolution_param {
+        num_output: 4116 #8*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+#--------------position sensitive RoI pooling--------------
+layer {
+    bottom: "rfcn_cls"
+    bottom: "rois"
+    top: "psroipooled_cls_rois"
+    name: "psroipooled_cls_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 21
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_cls_rois"
+    top: "cls_score"
+    name: "ave_cls_score_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+layer {
+    bottom: "rfcn_bbox"
+    bottom: "rois"
+    top: "psroipooled_loc_rois"
+    name: "psroipooled_loc_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 84
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_loc_rois"
+    top: "bbox_pred"
+    name: "ave_bbox_pred_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+#--------------online hard example mining--------------
+layer {
+   name: "per_roi_loss_cls"
+   type: "SoftmaxWithLossOHEM"
+   bottom: "cls_score"
+   bottom: "labels"
+   top: "temp_loss_cls"
+   top: "temp_prob_cls"
+   top: "per_roi_loss_cls"
+   loss_weight: 0
+   loss_weight: 0
+   loss_weight: 0
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "per_roi_loss_bbox"
+   type: "SmoothL1LossOHEM"
+   bottom: "bbox_pred"
+   bottom: "bbox_targets"
+   bottom: "bbox_inside_weights"
+   top: "temp_loss_bbox"
+   top: "per_roi_loss_bbox"
+   loss_weight: 0
+   loss_weight: 0
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "per_roi_loss"
+   type: "Eltwise"
+   bottom: "per_roi_loss_cls"
+   bottom: "per_roi_loss_bbox"
+   top: "per_roi_loss"
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   bottom: "rois"
+   bottom: "per_roi_loss"
+   bottom: "labels"
+   bottom: "bbox_inside_weights"
+   top: "labels_ohem"
+   top: "bbox_loss_weights_ohem"
+   name: "annotator_detector"
+   type: "BoxAnnotatorOHEM"
+   box_annotator_ohem_param {
+        roi_per_img: 128
+        ignore_label: -1
+   }
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "silence"
+   type: "Silence"
+   bottom: "bbox_outside_weights"
+   bottom: "temp_loss_cls"
+   bottom: "temp_prob_cls"
+   bottom: "temp_loss_bbox"
+}
+
+#-----------------------output------------------------
+layer {
+   name: "loss"
+   type: "SoftmaxWithLoss"
+   bottom: "cls_score"
+   bottom: "labels_ohem"
+   top: "loss_cls"
+   loss_weight: 1
+   loss_param {
+        ignore_label: -1
+   }
+   propagate_down: true
+   propagate_down: false
+}
+
+layer {
+   name: "accuarcy"
+   type: "Accuracy"
+   bottom: "cls_score"
+   bottom: "labels_ohem"
+   top: "accuarcy"
+   #include: { phase: TEST }
+   accuracy_param {
+        ignore_label: -1
+   }
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "loss_bbox"
+   type: "SmoothL1LossOHEM"
+   bottom: "bbox_pred"
+   bottom: "bbox_targets"
+   bottom: "bbox_loss_weights_ohem"
+   top: "loss_bbox"
+   loss_weight: 1
+   loss_param {
+        normalization: PRE_FIXED
+        pre_fixed_normalizer: 128
+   }
+   propagate_down: true
+   propagate_down: false
+   propagate_down: false
+}
+
diff --git a/models/pascal_voc/ResNet-101/rfcn_end2end/solver.prototxt b/models/pascal_voc/ResNet-101/rfcn_end2end/solver.prototxt
new file mode 100644
index 0000000..40e0173
--- /dev/null
+++ b/models/pascal_voc/ResNet-101/rfcn_end2end/solver.prototxt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 80000
+display: 20
+
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "resnet101_rfcn"
+iter_size: 2
+# debug_info: true
diff --git a/models/pascal_voc/ResNet-101/rfcn_end2end/solver_ohem.prototxt b/models/pascal_voc/ResNet-101/rfcn_end2end/solver_ohem.prototxt
new file mode 100644
index 0000000..5247534
--- /dev/null
+++ b/models/pascal_voc/ResNet-101/rfcn_end2end/solver_ohem.prototxt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic_ohem.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 80000
+display: 20
+
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "resnet101_rfcn_ohem"
+iter_size: 2
+# debug_info: true
diff --git a/models/pascal_voc/ResNet-101/rfcn_end2end/solver_ohem_continue.prototxt~ b/models/pascal_voc/ResNet-101/rfcn_end2end/solver_ohem_continue.prototxt~
new file mode 100644
index 0000000..e4b9335
--- /dev/null
+++ b/models/pascal_voc/ResNet-101/rfcn_end2end/solver_ohem_continue.prototxt~
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic_ohem.prototxt"
+base_lr: 0.001
+lr_policy: "multistep"
+gamma: 0.1
+stepvalue: 10000
+display: 20
+
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "resnet101_rfcn_ohem"
+iter_size: 2
+# debug_info: true
diff --git a/models/pascal_voc/ResNet-101/rfcn_end2end/solver_warmup.prototxt b/models/pascal_voc/ResNet-101/rfcn_end2end/solver_warmup.prototxt
new file mode 100644
index 0000000..39e5912
--- /dev/null
+++ b/models/pascal_voc/ResNet-101/rfcn_end2end/solver_warmup.prototxt
@@ -0,0 +1,14 @@
+train_net: "models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic_ohem.prototxt"
+lr_policy: "fixed"
+base_lr: 0.0001
+
+display: 20
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "resnet101_rfcn_ohem"
+iter_size: 2
+# debug_info: true
diff --git a/models/pascal_voc/ResNet-101/rfcn_end2end/solver_warmup_continue.prototxt b/models/pascal_voc/ResNet-101/rfcn_end2end/solver_warmup_continue.prototxt
new file mode 100644
index 0000000..06d470c
--- /dev/null
+++ b/models/pascal_voc/ResNet-101/rfcn_end2end/solver_warmup_continue.prototxt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic_ohem.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 70000
+
+display: 20
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "resnet101_rfcn_ohem"
+iter_size: 2
+# debug_info: true
diff --git a/models/pascal_voc/ResNet-101/rfcn_end2end/test_agonistic.prototxt b/models/pascal_voc/ResNet-101/rfcn_end2end/test_agonistic.prototxt
new file mode 100644
index 0000000..a2b0965
--- /dev/null
+++ b/models/pascal_voc/ResNet-101/rfcn_end2end/test_agonistic.prototxt
@@ -0,0 +1,7186 @@
+name: "ResNet-101"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+# ------------------------ conv1 -----------------------------
+layer {
+    bottom: "data"
+    top: "conv1"
+    name: "conv1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 7
+        pad: 3
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "bn_conv1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "scale_conv1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "conv1"
+    bottom: "conv1"
+    name: "conv1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv1"
+    top: "pool1"
+    name: "pool1"
+    type: "Pooling"
+    pooling_param {
+        kernel_size: 3
+        stride: 2
+        pool: MAX
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch1"
+    name: "res2a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "bn2a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "scale2a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "bn2a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "scale2a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2a_branch2a"
+    bottom: "res2a_branch2a"
+    name: "res2a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "bn2a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "scale2a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2a_branch2b"
+    bottom: "res2a_branch2b"
+    name: "res2a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2c"
+    name: "res2a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "bn2a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "scale2a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    bottom: "res2a_branch2c"
+    top: "res2a"
+    name: "res2a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2a"
+    name: "res2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "bn2b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "scale2b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2b_branch2a"
+    bottom: "res2b_branch2a"
+    name: "res2b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "bn2b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "scale2b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2b_branch2b"
+    bottom: "res2b_branch2b"
+    name: "res2b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2c"
+    name: "res2b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "bn2b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "scale2b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a"
+    bottom: "res2b_branch2c"
+    top: "res2b"
+    name: "res2b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2b"
+    name: "res2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "bn2c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "scale2c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2c_branch2a"
+    bottom: "res2c_branch2a"
+    name: "res2c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "bn2c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "scale2c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2c_branch2b"
+    bottom: "res2c_branch2b"
+    name: "res2c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2c"
+    name: "res2c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "bn2c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "scale2c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b"
+    bottom: "res2c_branch2c"
+    top: "res2c"
+    name: "res2c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res2c"
+    name: "res2c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch1"
+    name: "res3a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "bn3a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "scale3a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "bn3a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "scale3a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3a_branch2a"
+    bottom: "res3a_branch2a"
+    name: "res3a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "bn3a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "scale3a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3a_branch2b"
+    bottom: "res3a_branch2b"
+    name: "res3a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2c"
+    name: "res3a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "bn3a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "scale3a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    bottom: "res3a_branch2c"
+    top: "res3a"
+    name: "res3a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3a"
+    name: "res3a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3b1_branch2a"
+    name: "res3b1_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2a"
+    name: "bn3b1_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2a"
+    name: "scale3b1_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b1_branch2a"
+    bottom: "res3b1_branch2a"
+    name: "res3b1_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2b"
+    name: "res3b1_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2b"
+    name: "bn3b1_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2b"
+    name: "scale3b1_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b1_branch2b"
+    bottom: "res3b1_branch2b"
+    name: "res3b1_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2c"
+    name: "res3b1_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2c"
+    top: "res3b1_branch2c"
+    name: "bn3b1_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2c"
+    top: "res3b1_branch2c"
+    name: "scale3b1_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a"
+    bottom: "res3b1_branch2c"
+    top: "res3b1"
+    name: "res3b1"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b1"
+    top: "res3b1"
+    name: "res3b1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1"
+    top: "res3b2_branch2a"
+    name: "res3b2_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2a"
+    name: "bn3b2_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2a"
+    name: "scale3b2_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b2_branch2a"
+    bottom: "res3b2_branch2a"
+    name: "res3b2_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2b"
+    name: "res3b2_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2b"
+    name: "bn3b2_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2b"
+    name: "scale3b2_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b2_branch2b"
+    bottom: "res3b2_branch2b"
+    name: "res3b2_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2c"
+    name: "res3b2_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2c"
+    top: "res3b2_branch2c"
+    name: "bn3b2_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2c"
+    top: "res3b2_branch2c"
+    name: "scale3b2_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1"
+    bottom: "res3b2_branch2c"
+    top: "res3b2"
+    name: "res3b2"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b2"
+    top: "res3b2"
+    name: "res3b2_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2"
+    top: "res3b3_branch2a"
+    name: "res3b3_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2a"
+    name: "bn3b3_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2a"
+    name: "scale3b3_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b3_branch2a"
+    bottom: "res3b3_branch2a"
+    name: "res3b3_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2b"
+    name: "res3b3_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2b"
+    name: "bn3b3_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2b"
+    name: "scale3b3_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b3_branch2b"
+    bottom: "res3b3_branch2b"
+    name: "res3b3_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2c"
+    name: "res3b3_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2c"
+    top: "res3b3_branch2c"
+    name: "bn3b3_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2c"
+    top: "res3b3_branch2c"
+    name: "scale3b3_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2"
+    bottom: "res3b3_branch2c"
+    top: "res3b3"
+    name: "res3b3"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res3b3"
+    name: "res3b3_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res4a_branch1"
+    name: "res4a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "bn4a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "scale4a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "bn4a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "scale4a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4a_branch2a"
+    bottom: "res4a_branch2a"
+    name: "res4a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "bn4a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "scale4a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4a_branch2b"
+    bottom: "res4a_branch2b"
+    name: "res4a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2c"
+    name: "res4a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "bn4a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "scale4a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    bottom: "res4a_branch2c"
+    top: "res4a"
+    name: "res4a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4a"
+    name: "res4a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4b1_branch2a"
+    name: "res4b1_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2a"
+    name: "bn4b1_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2a"
+    name: "scale4b1_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b1_branch2a"
+    bottom: "res4b1_branch2a"
+    name: "res4b1_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2b"
+    name: "res4b1_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2b"
+    name: "bn4b1_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2b"
+    name: "scale4b1_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b1_branch2b"
+    bottom: "res4b1_branch2b"
+    name: "res4b1_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2c"
+    name: "res4b1_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2c"
+    top: "res4b1_branch2c"
+    name: "bn4b1_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2c"
+    top: "res4b1_branch2c"
+    name: "scale4b1_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a"
+    bottom: "res4b1_branch2c"
+    top: "res4b1"
+    name: "res4b1"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b1"
+    top: "res4b1"
+    name: "res4b1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1"
+    top: "res4b2_branch2a"
+    name: "res4b2_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2a"
+    name: "bn4b2_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2a"
+    name: "scale4b2_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b2_branch2a"
+    bottom: "res4b2_branch2a"
+    name: "res4b2_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2b"
+    name: "res4b2_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2b"
+    name: "bn4b2_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2b"
+    name: "scale4b2_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b2_branch2b"
+    bottom: "res4b2_branch2b"
+    name: "res4b2_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2c"
+    name: "res4b2_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2c"
+    top: "res4b2_branch2c"
+    name: "bn4b2_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2c"
+    top: "res4b2_branch2c"
+    name: "scale4b2_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1"
+    bottom: "res4b2_branch2c"
+    top: "res4b2"
+    name: "res4b2"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b2"
+    top: "res4b2"
+    name: "res4b2_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2"
+    top: "res4b3_branch2a"
+    name: "res4b3_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2a"
+    name: "bn4b3_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2a"
+    name: "scale4b3_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b3_branch2a"
+    bottom: "res4b3_branch2a"
+    name: "res4b3_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2b"
+    name: "res4b3_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2b"
+    name: "bn4b3_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2b"
+    name: "scale4b3_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b3_branch2b"
+    bottom: "res4b3_branch2b"
+    name: "res4b3_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2c"
+    name: "res4b3_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2c"
+    top: "res4b3_branch2c"
+    name: "bn4b3_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2c"
+    top: "res4b3_branch2c"
+    name: "scale4b3_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2"
+    bottom: "res4b3_branch2c"
+    top: "res4b3"
+    name: "res4b3"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b3"
+    top: "res4b3"
+    name: "res4b3_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3"
+    top: "res4b4_branch2a"
+    name: "res4b4_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2a"
+    name: "bn4b4_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2a"
+    name: "scale4b4_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b4_branch2a"
+    bottom: "res4b4_branch2a"
+    name: "res4b4_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2b"
+    name: "res4b4_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2b"
+    name: "bn4b4_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2b"
+    name: "scale4b4_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b4_branch2b"
+    bottom: "res4b4_branch2b"
+    name: "res4b4_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2c"
+    name: "res4b4_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2c"
+    top: "res4b4_branch2c"
+    name: "bn4b4_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2c"
+    top: "res4b4_branch2c"
+    name: "scale4b4_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3"
+    bottom: "res4b4_branch2c"
+    top: "res4b4"
+    name: "res4b4"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b4"
+    top: "res4b4"
+    name: "res4b4_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4"
+    top: "res4b5_branch2a"
+    name: "res4b5_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2a"
+    name: "bn4b5_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2a"
+    name: "scale4b5_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b5_branch2a"
+    bottom: "res4b5_branch2a"
+    name: "res4b5_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2b"
+    name: "res4b5_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2b"
+    name: "bn4b5_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2b"
+    name: "scale4b5_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b5_branch2b"
+    bottom: "res4b5_branch2b"
+    name: "res4b5_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2c"
+    name: "res4b5_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2c"
+    top: "res4b5_branch2c"
+    name: "bn4b5_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2c"
+    top: "res4b5_branch2c"
+    name: "scale4b5_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4"
+    bottom: "res4b5_branch2c"
+    top: "res4b5"
+    name: "res4b5"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b5"
+    top: "res4b5"
+    name: "res4b5_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5"
+    top: "res4b6_branch2a"
+    name: "res4b6_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2a"
+    name: "bn4b6_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2a"
+    name: "scale4b6_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b6_branch2a"
+    bottom: "res4b6_branch2a"
+    name: "res4b6_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2b"
+    name: "res4b6_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2b"
+    name: "bn4b6_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2b"
+    name: "scale4b6_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b6_branch2b"
+    bottom: "res4b6_branch2b"
+    name: "res4b6_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2c"
+    name: "res4b6_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2c"
+    top: "res4b6_branch2c"
+    name: "bn4b6_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2c"
+    top: "res4b6_branch2c"
+    name: "scale4b6_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5"
+    bottom: "res4b6_branch2c"
+    top: "res4b6"
+    name: "res4b6"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b6"
+    top: "res4b6"
+    name: "res4b6_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6"
+    top: "res4b7_branch2a"
+    name: "res4b7_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2a"
+    name: "bn4b7_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2a"
+    name: "scale4b7_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b7_branch2a"
+    bottom: "res4b7_branch2a"
+    name: "res4b7_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2b"
+    name: "res4b7_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2b"
+    name: "bn4b7_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2b"
+    name: "scale4b7_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b7_branch2b"
+    bottom: "res4b7_branch2b"
+    name: "res4b7_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2c"
+    name: "res4b7_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2c"
+    top: "res4b7_branch2c"
+    name: "bn4b7_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2c"
+    top: "res4b7_branch2c"
+    name: "scale4b7_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6"
+    bottom: "res4b7_branch2c"
+    top: "res4b7"
+    name: "res4b7"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b7"
+    top: "res4b7"
+    name: "res4b7_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7"
+    top: "res4b8_branch2a"
+    name: "res4b8_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2a"
+    name: "bn4b8_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2a"
+    name: "scale4b8_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b8_branch2a"
+    bottom: "res4b8_branch2a"
+    name: "res4b8_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2b"
+    name: "res4b8_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2b"
+    name: "bn4b8_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2b"
+    name: "scale4b8_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b8_branch2b"
+    bottom: "res4b8_branch2b"
+    name: "res4b8_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2c"
+    name: "res4b8_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2c"
+    top: "res4b8_branch2c"
+    name: "bn4b8_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2c"
+    top: "res4b8_branch2c"
+    name: "scale4b8_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7"
+    bottom: "res4b8_branch2c"
+    top: "res4b8"
+    name: "res4b8"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b8"
+    top: "res4b8"
+    name: "res4b8_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8"
+    top: "res4b9_branch2a"
+    name: "res4b9_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2a"
+    name: "bn4b9_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2a"
+    name: "scale4b9_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b9_branch2a"
+    bottom: "res4b9_branch2a"
+    name: "res4b9_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2b"
+    name: "res4b9_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2b"
+    name: "bn4b9_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2b"
+    name: "scale4b9_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b9_branch2b"
+    bottom: "res4b9_branch2b"
+    name: "res4b9_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2c"
+    name: "res4b9_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2c"
+    top: "res4b9_branch2c"
+    name: "bn4b9_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2c"
+    top: "res4b9_branch2c"
+    name: "scale4b9_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8"
+    bottom: "res4b9_branch2c"
+    top: "res4b9"
+    name: "res4b9"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b9"
+    top: "res4b9"
+    name: "res4b9_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9"
+    top: "res4b10_branch2a"
+    name: "res4b10_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2a"
+    name: "bn4b10_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2a"
+    name: "scale4b10_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b10_branch2a"
+    bottom: "res4b10_branch2a"
+    name: "res4b10_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2b"
+    name: "res4b10_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2b"
+    name: "bn4b10_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2b"
+    name: "scale4b10_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b10_branch2b"
+    bottom: "res4b10_branch2b"
+    name: "res4b10_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2c"
+    name: "res4b10_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2c"
+    top: "res4b10_branch2c"
+    name: "bn4b10_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2c"
+    top: "res4b10_branch2c"
+    name: "scale4b10_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9"
+    bottom: "res4b10_branch2c"
+    top: "res4b10"
+    name: "res4b10"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b10"
+    top: "res4b10"
+    name: "res4b10_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10"
+    top: "res4b11_branch2a"
+    name: "res4b11_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2a"
+    name: "bn4b11_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2a"
+    name: "scale4b11_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b11_branch2a"
+    bottom: "res4b11_branch2a"
+    name: "res4b11_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2b"
+    name: "res4b11_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2b"
+    name: "bn4b11_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2b"
+    name: "scale4b11_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b11_branch2b"
+    bottom: "res4b11_branch2b"
+    name: "res4b11_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2c"
+    name: "res4b11_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2c"
+    top: "res4b11_branch2c"
+    name: "bn4b11_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2c"
+    top: "res4b11_branch2c"
+    name: "scale4b11_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10"
+    bottom: "res4b11_branch2c"
+    top: "res4b11"
+    name: "res4b11"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b11"
+    top: "res4b11"
+    name: "res4b11_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11"
+    top: "res4b12_branch2a"
+    name: "res4b12_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2a"
+    name: "bn4b12_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2a"
+    name: "scale4b12_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b12_branch2a"
+    bottom: "res4b12_branch2a"
+    name: "res4b12_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2b"
+    name: "res4b12_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2b"
+    name: "bn4b12_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2b"
+    name: "scale4b12_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b12_branch2b"
+    bottom: "res4b12_branch2b"
+    name: "res4b12_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2c"
+    name: "res4b12_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2c"
+    top: "res4b12_branch2c"
+    name: "bn4b12_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2c"
+    top: "res4b12_branch2c"
+    name: "scale4b12_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11"
+    bottom: "res4b12_branch2c"
+    top: "res4b12"
+    name: "res4b12"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b12"
+    top: "res4b12"
+    name: "res4b12_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12"
+    top: "res4b13_branch2a"
+    name: "res4b13_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2a"
+    name: "bn4b13_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2a"
+    name: "scale4b13_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b13_branch2a"
+    bottom: "res4b13_branch2a"
+    name: "res4b13_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2b"
+    name: "res4b13_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2b"
+    name: "bn4b13_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2b"
+    name: "scale4b13_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b13_branch2b"
+    bottom: "res4b13_branch2b"
+    name: "res4b13_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2c"
+    name: "res4b13_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2c"
+    top: "res4b13_branch2c"
+    name: "bn4b13_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2c"
+    top: "res4b13_branch2c"
+    name: "scale4b13_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12"
+    bottom: "res4b13_branch2c"
+    top: "res4b13"
+    name: "res4b13"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b13"
+    top: "res4b13"
+    name: "res4b13_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13"
+    top: "res4b14_branch2a"
+    name: "res4b14_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2a"
+    name: "bn4b14_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2a"
+    name: "scale4b14_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b14_branch2a"
+    bottom: "res4b14_branch2a"
+    name: "res4b14_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2b"
+    name: "res4b14_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2b"
+    name: "bn4b14_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2b"
+    name: "scale4b14_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b14_branch2b"
+    bottom: "res4b14_branch2b"
+    name: "res4b14_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2c"
+    name: "res4b14_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2c"
+    top: "res4b14_branch2c"
+    name: "bn4b14_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2c"
+    top: "res4b14_branch2c"
+    name: "scale4b14_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13"
+    bottom: "res4b14_branch2c"
+    top: "res4b14"
+    name: "res4b14"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b14"
+    top: "res4b14"
+    name: "res4b14_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14"
+    top: "res4b15_branch2a"
+    name: "res4b15_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2a"
+    name: "bn4b15_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2a"
+    name: "scale4b15_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b15_branch2a"
+    bottom: "res4b15_branch2a"
+    name: "res4b15_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2b"
+    name: "res4b15_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2b"
+    name: "bn4b15_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2b"
+    name: "scale4b15_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b15_branch2b"
+    bottom: "res4b15_branch2b"
+    name: "res4b15_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2c"
+    name: "res4b15_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2c"
+    top: "res4b15_branch2c"
+    name: "bn4b15_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2c"
+    top: "res4b15_branch2c"
+    name: "scale4b15_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14"
+    bottom: "res4b15_branch2c"
+    top: "res4b15"
+    name: "res4b15"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b15"
+    top: "res4b15"
+    name: "res4b15_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15"
+    top: "res4b16_branch2a"
+    name: "res4b16_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2a"
+    name: "bn4b16_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2a"
+    name: "scale4b16_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b16_branch2a"
+    bottom: "res4b16_branch2a"
+    name: "res4b16_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2b"
+    name: "res4b16_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2b"
+    name: "bn4b16_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2b"
+    name: "scale4b16_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b16_branch2b"
+    bottom: "res4b16_branch2b"
+    name: "res4b16_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2c"
+    name: "res4b16_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2c"
+    top: "res4b16_branch2c"
+    name: "bn4b16_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2c"
+    top: "res4b16_branch2c"
+    name: "scale4b16_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15"
+    bottom: "res4b16_branch2c"
+    top: "res4b16"
+    name: "res4b16"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b16"
+    top: "res4b16"
+    name: "res4b16_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16"
+    top: "res4b17_branch2a"
+    name: "res4b17_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2a"
+    name: "bn4b17_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2a"
+    name: "scale4b17_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b17_branch2a"
+    bottom: "res4b17_branch2a"
+    name: "res4b17_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2b"
+    name: "res4b17_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2b"
+    name: "bn4b17_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2b"
+    name: "scale4b17_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b17_branch2b"
+    bottom: "res4b17_branch2b"
+    name: "res4b17_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2c"
+    name: "res4b17_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2c"
+    top: "res4b17_branch2c"
+    name: "bn4b17_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2c"
+    top: "res4b17_branch2c"
+    name: "scale4b17_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16"
+    bottom: "res4b17_branch2c"
+    top: "res4b17"
+    name: "res4b17"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b17"
+    top: "res4b17"
+    name: "res4b17_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17"
+    top: "res4b18_branch2a"
+    name: "res4b18_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2a"
+    name: "bn4b18_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2a"
+    name: "scale4b18_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b18_branch2a"
+    bottom: "res4b18_branch2a"
+    name: "res4b18_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2b"
+    name: "res4b18_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2b"
+    name: "bn4b18_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2b"
+    name: "scale4b18_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b18_branch2b"
+    bottom: "res4b18_branch2b"
+    name: "res4b18_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2c"
+    name: "res4b18_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2c"
+    top: "res4b18_branch2c"
+    name: "bn4b18_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2c"
+    top: "res4b18_branch2c"
+    name: "scale4b18_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17"
+    bottom: "res4b18_branch2c"
+    top: "res4b18"
+    name: "res4b18"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b18"
+    top: "res4b18"
+    name: "res4b18_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18"
+    top: "res4b19_branch2a"
+    name: "res4b19_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2a"
+    name: "bn4b19_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2a"
+    name: "scale4b19_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b19_branch2a"
+    bottom: "res4b19_branch2a"
+    name: "res4b19_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2b"
+    name: "res4b19_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2b"
+    name: "bn4b19_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2b"
+    name: "scale4b19_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b19_branch2b"
+    bottom: "res4b19_branch2b"
+    name: "res4b19_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2c"
+    name: "res4b19_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2c"
+    top: "res4b19_branch2c"
+    name: "bn4b19_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2c"
+    top: "res4b19_branch2c"
+    name: "scale4b19_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18"
+    bottom: "res4b19_branch2c"
+    top: "res4b19"
+    name: "res4b19"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b19"
+    top: "res4b19"
+    name: "res4b19_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19"
+    top: "res4b20_branch2a"
+    name: "res4b20_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2a"
+    name: "bn4b20_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2a"
+    name: "scale4b20_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b20_branch2a"
+    bottom: "res4b20_branch2a"
+    name: "res4b20_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2b"
+    name: "res4b20_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2b"
+    name: "bn4b20_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2b"
+    name: "scale4b20_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b20_branch2b"
+    bottom: "res4b20_branch2b"
+    name: "res4b20_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2c"
+    name: "res4b20_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2c"
+    top: "res4b20_branch2c"
+    name: "bn4b20_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2c"
+    top: "res4b20_branch2c"
+    name: "scale4b20_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19"
+    bottom: "res4b20_branch2c"
+    top: "res4b20"
+    name: "res4b20"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b20"
+    top: "res4b20"
+    name: "res4b20_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20"
+    top: "res4b21_branch2a"
+    name: "res4b21_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2a"
+    name: "bn4b21_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2a"
+    name: "scale4b21_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b21_branch2a"
+    bottom: "res4b21_branch2a"
+    name: "res4b21_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2b"
+    name: "res4b21_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2b"
+    name: "bn4b21_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2b"
+    name: "scale4b21_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b21_branch2b"
+    bottom: "res4b21_branch2b"
+    name: "res4b21_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2c"
+    name: "res4b21_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2c"
+    top: "res4b21_branch2c"
+    name: "bn4b21_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2c"
+    top: "res4b21_branch2c"
+    name: "scale4b21_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20"
+    bottom: "res4b21_branch2c"
+    top: "res4b21"
+    name: "res4b21"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b21"
+    top: "res4b21"
+    name: "res4b21_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21"
+    top: "res4b22_branch2a"
+    name: "res4b22_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2a"
+    name: "bn4b22_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2a"
+    name: "scale4b22_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b22_branch2a"
+    bottom: "res4b22_branch2a"
+    name: "res4b22_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2b"
+    name: "res4b22_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2b"
+    name: "bn4b22_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2b"
+    name: "scale4b22_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b22_branch2b"
+    bottom: "res4b22_branch2b"
+    name: "res4b22_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2c"
+    name: "res4b22_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2c"
+    top: "res4b22_branch2c"
+    name: "bn4b22_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2c"
+    top: "res4b22_branch2c"
+    name: "scale4b22_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21"
+    bottom: "res4b22_branch2c"
+    top: "res4b22"
+    name: "res4b22"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res4b22"
+    name: "res4b22_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res5a_branch1"
+    name: "res5a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "bn5a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "scale5a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "bn5a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "scale5a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5a_branch2a"
+    bottom: "res5a_branch2a"
+    name: "res5a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "bn5a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "scale5a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5a_branch2b"
+    bottom: "res5a_branch2b"
+    name: "res5a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2c"
+    name: "res5a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "bn5a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "scale5a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    bottom: "res5a_branch2c"
+    top: "res5a"
+    name: "res5a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5a"
+    name: "res5a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "bn5b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "scale5b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5b_branch2a"
+    bottom: "res5b_branch2a"
+    name: "res5b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "bn5b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "scale5b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5b_branch2b"
+    bottom: "res5b_branch2b"
+    name: "res5b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2c"
+    name: "res5b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "bn5b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "scale5b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a"
+    bottom: "res5b_branch2c"
+    top: "res5b"
+    name: "res5b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5b"
+    name: "res5b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "bn5c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "scale5c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5c_branch2a"
+    bottom: "res5c_branch2a"
+    name: "res5c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "bn5c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "scale5c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5c_branch2b"
+    bottom: "res5c_branch2b"
+    name: "res5c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2c"
+    name: "res5c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "bn5c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "scale5c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b"
+    bottom: "res5c_branch2c"
+    top: "res5c"
+    name: "res5c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5c"
+    top: "res5c"
+    name: "res5c_relu"
+    type: "ReLU"
+}
+
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "res5c"
+  top: "rpn/output"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#----------------------new conv layer------------------
+layer {
+    bottom: "res5c"
+    top: "conv_new_1"
+    name: "conv_new_1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "conv_new_1"
+    name: "conv_new_1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_cls"
+    name: "rfcn_cls"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1029 #21*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_bbox"
+    name: "rfcn_bbox"
+    type: "Convolution"
+    convolution_param {
+        num_output: 392 #8*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+#--------------position sensitive RoI pooling--------------
+layer {
+    bottom: "rfcn_cls"
+    bottom: "rois"
+    top: "psroipooled_cls_rois"
+    name: "psroipooled_cls_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 21
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_cls_rois"
+    top: "cls_score"
+    name: "ave_cls_score_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+layer {
+    bottom: "rfcn_bbox"
+    bottom: "rois"
+    top: "psroipooled_loc_rois"
+    name: "psroipooled_loc_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 8
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_loc_rois"
+    top: "bbox_pred_pre"
+    name: "ave_bbox_pred_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+#-----------------------output------------------------
+layer {
+   name: "cls_prob"
+   type: "Softmax"
+   bottom: "cls_score"
+   top: "cls_prob_pre"
+}
+
+layer {
+    name: "cls_prob_reshape"
+    type: "Reshape"
+    bottom: "cls_prob_pre"
+    top: "cls_prob"
+    reshape_param {
+        shape {
+            dim: -1
+            dim: 21
+        }
+    }
+}
+
+layer {
+    name: "bbox_pred_reshape"
+    type: "Reshape"
+    bottom: "bbox_pred_pre"
+    top: "bbox_pred"
+    reshape_param {
+        shape {
+            dim: -1
+            dim: 8
+        }
+    }
+}
+
+
diff --git a/models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic.prototxt b/models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic.prototxt
new file mode 100644
index 0000000..4580c66
--- /dev/null
+++ b/models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic.prototxt
@@ -0,0 +1,7274 @@
+name: "ResNet-101"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+# ------------------------ conv1 -----------------------------
+layer {
+    bottom: "data"
+    top: "conv1"
+    name: "conv1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 7
+        pad: 3
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "bn_conv1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "scale_conv1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "conv1"
+    bottom: "conv1"
+    name: "conv1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv1"
+    top: "pool1"
+    name: "pool1"
+    type: "Pooling"
+    pooling_param {
+        kernel_size: 3
+        stride: 2
+        pool: MAX
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch1"
+    name: "res2a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "bn2a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "scale2a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "bn2a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "scale2a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2a_branch2a"
+    bottom: "res2a_branch2a"
+    name: "res2a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "bn2a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "scale2a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2a_branch2b"
+    bottom: "res2a_branch2b"
+    name: "res2a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2c"
+    name: "res2a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "bn2a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "scale2a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    bottom: "res2a_branch2c"
+    top: "res2a"
+    name: "res2a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2a"
+    name: "res2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "bn2b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "scale2b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2b_branch2a"
+    bottom: "res2b_branch2a"
+    name: "res2b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "bn2b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "scale2b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2b_branch2b"
+    bottom: "res2b_branch2b"
+    name: "res2b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2c"
+    name: "res2b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "bn2b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "scale2b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a"
+    bottom: "res2b_branch2c"
+    top: "res2b"
+    name: "res2b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2b"
+    name: "res2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "bn2c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "scale2c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2c_branch2a"
+    bottom: "res2c_branch2a"
+    name: "res2c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "bn2c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "scale2c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2c_branch2b"
+    bottom: "res2c_branch2b"
+    name: "res2c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2c"
+    name: "res2c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "bn2c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "scale2c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b"
+    bottom: "res2c_branch2c"
+    top: "res2c"
+    name: "res2c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res2c"
+    name: "res2c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch1"
+    name: "res3a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "bn3a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "scale3a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "bn3a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "scale3a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3a_branch2a"
+    bottom: "res3a_branch2a"
+    name: "res3a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "bn3a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "scale3a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3a_branch2b"
+    bottom: "res3a_branch2b"
+    name: "res3a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2c"
+    name: "res3a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "bn3a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "scale3a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    bottom: "res3a_branch2c"
+    top: "res3a"
+    name: "res3a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3a"
+    name: "res3a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3b1_branch2a"
+    name: "res3b1_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2a"
+    name: "bn3b1_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2a"
+    name: "scale3b1_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b1_branch2a"
+    bottom: "res3b1_branch2a"
+    name: "res3b1_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2b"
+    name: "res3b1_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2b"
+    name: "bn3b1_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2b"
+    name: "scale3b1_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b1_branch2b"
+    bottom: "res3b1_branch2b"
+    name: "res3b1_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2c"
+    name: "res3b1_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2c"
+    top: "res3b1_branch2c"
+    name: "bn3b1_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2c"
+    top: "res3b1_branch2c"
+    name: "scale3b1_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a"
+    bottom: "res3b1_branch2c"
+    top: "res3b1"
+    name: "res3b1"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b1"
+    top: "res3b1"
+    name: "res3b1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1"
+    top: "res3b2_branch2a"
+    name: "res3b2_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2a"
+    name: "bn3b2_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2a"
+    name: "scale3b2_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b2_branch2a"
+    bottom: "res3b2_branch2a"
+    name: "res3b2_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2b"
+    name: "res3b2_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2b"
+    name: "bn3b2_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2b"
+    name: "scale3b2_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b2_branch2b"
+    bottom: "res3b2_branch2b"
+    name: "res3b2_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2c"
+    name: "res3b2_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2c"
+    top: "res3b2_branch2c"
+    name: "bn3b2_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2c"
+    top: "res3b2_branch2c"
+    name: "scale3b2_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1"
+    bottom: "res3b2_branch2c"
+    top: "res3b2"
+    name: "res3b2"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b2"
+    top: "res3b2"
+    name: "res3b2_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2"
+    top: "res3b3_branch2a"
+    name: "res3b3_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2a"
+    name: "bn3b3_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2a"
+    name: "scale3b3_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b3_branch2a"
+    bottom: "res3b3_branch2a"
+    name: "res3b3_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2b"
+    name: "res3b3_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2b"
+    name: "bn3b3_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2b"
+    name: "scale3b3_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b3_branch2b"
+    bottom: "res3b3_branch2b"
+    name: "res3b3_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2c"
+    name: "res3b3_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2c"
+    top: "res3b3_branch2c"
+    name: "bn3b3_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2c"
+    top: "res3b3_branch2c"
+    name: "scale3b3_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2"
+    bottom: "res3b3_branch2c"
+    top: "res3b3"
+    name: "res3b3"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res3b3"
+    name: "res3b3_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res4a_branch1"
+    name: "res4a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "bn4a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "scale4a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "bn4a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "scale4a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4a_branch2a"
+    bottom: "res4a_branch2a"
+    name: "res4a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "bn4a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "scale4a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4a_branch2b"
+    bottom: "res4a_branch2b"
+    name: "res4a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2c"
+    name: "res4a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "bn4a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "scale4a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    bottom: "res4a_branch2c"
+    top: "res4a"
+    name: "res4a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4a"
+    name: "res4a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4b1_branch2a"
+    name: "res4b1_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2a"
+    name: "bn4b1_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2a"
+    name: "scale4b1_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b1_branch2a"
+    bottom: "res4b1_branch2a"
+    name: "res4b1_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2b"
+    name: "res4b1_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2b"
+    name: "bn4b1_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2b"
+    name: "scale4b1_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b1_branch2b"
+    bottom: "res4b1_branch2b"
+    name: "res4b1_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2c"
+    name: "res4b1_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2c"
+    top: "res4b1_branch2c"
+    name: "bn4b1_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2c"
+    top: "res4b1_branch2c"
+    name: "scale4b1_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a"
+    bottom: "res4b1_branch2c"
+    top: "res4b1"
+    name: "res4b1"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b1"
+    top: "res4b1"
+    name: "res4b1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1"
+    top: "res4b2_branch2a"
+    name: "res4b2_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2a"
+    name: "bn4b2_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2a"
+    name: "scale4b2_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b2_branch2a"
+    bottom: "res4b2_branch2a"
+    name: "res4b2_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2b"
+    name: "res4b2_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2b"
+    name: "bn4b2_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2b"
+    name: "scale4b2_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b2_branch2b"
+    bottom: "res4b2_branch2b"
+    name: "res4b2_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2c"
+    name: "res4b2_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2c"
+    top: "res4b2_branch2c"
+    name: "bn4b2_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2c"
+    top: "res4b2_branch2c"
+    name: "scale4b2_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1"
+    bottom: "res4b2_branch2c"
+    top: "res4b2"
+    name: "res4b2"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b2"
+    top: "res4b2"
+    name: "res4b2_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2"
+    top: "res4b3_branch2a"
+    name: "res4b3_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2a"
+    name: "bn4b3_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2a"
+    name: "scale4b3_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b3_branch2a"
+    bottom: "res4b3_branch2a"
+    name: "res4b3_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2b"
+    name: "res4b3_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2b"
+    name: "bn4b3_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2b"
+    name: "scale4b3_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b3_branch2b"
+    bottom: "res4b3_branch2b"
+    name: "res4b3_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2c"
+    name: "res4b3_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2c"
+    top: "res4b3_branch2c"
+    name: "bn4b3_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2c"
+    top: "res4b3_branch2c"
+    name: "scale4b3_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2"
+    bottom: "res4b3_branch2c"
+    top: "res4b3"
+    name: "res4b3"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b3"
+    top: "res4b3"
+    name: "res4b3_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3"
+    top: "res4b4_branch2a"
+    name: "res4b4_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2a"
+    name: "bn4b4_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2a"
+    name: "scale4b4_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b4_branch2a"
+    bottom: "res4b4_branch2a"
+    name: "res4b4_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2b"
+    name: "res4b4_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2b"
+    name: "bn4b4_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2b"
+    name: "scale4b4_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b4_branch2b"
+    bottom: "res4b4_branch2b"
+    name: "res4b4_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2c"
+    name: "res4b4_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2c"
+    top: "res4b4_branch2c"
+    name: "bn4b4_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2c"
+    top: "res4b4_branch2c"
+    name: "scale4b4_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3"
+    bottom: "res4b4_branch2c"
+    top: "res4b4"
+    name: "res4b4"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b4"
+    top: "res4b4"
+    name: "res4b4_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4"
+    top: "res4b5_branch2a"
+    name: "res4b5_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2a"
+    name: "bn4b5_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2a"
+    name: "scale4b5_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b5_branch2a"
+    bottom: "res4b5_branch2a"
+    name: "res4b5_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2b"
+    name: "res4b5_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2b"
+    name: "bn4b5_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2b"
+    name: "scale4b5_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b5_branch2b"
+    bottom: "res4b5_branch2b"
+    name: "res4b5_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2c"
+    name: "res4b5_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2c"
+    top: "res4b5_branch2c"
+    name: "bn4b5_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2c"
+    top: "res4b5_branch2c"
+    name: "scale4b5_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4"
+    bottom: "res4b5_branch2c"
+    top: "res4b5"
+    name: "res4b5"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b5"
+    top: "res4b5"
+    name: "res4b5_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5"
+    top: "res4b6_branch2a"
+    name: "res4b6_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2a"
+    name: "bn4b6_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2a"
+    name: "scale4b6_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b6_branch2a"
+    bottom: "res4b6_branch2a"
+    name: "res4b6_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2b"
+    name: "res4b6_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2b"
+    name: "bn4b6_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2b"
+    name: "scale4b6_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b6_branch2b"
+    bottom: "res4b6_branch2b"
+    name: "res4b6_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2c"
+    name: "res4b6_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2c"
+    top: "res4b6_branch2c"
+    name: "bn4b6_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2c"
+    top: "res4b6_branch2c"
+    name: "scale4b6_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5"
+    bottom: "res4b6_branch2c"
+    top: "res4b6"
+    name: "res4b6"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b6"
+    top: "res4b6"
+    name: "res4b6_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6"
+    top: "res4b7_branch2a"
+    name: "res4b7_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2a"
+    name: "bn4b7_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2a"
+    name: "scale4b7_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b7_branch2a"
+    bottom: "res4b7_branch2a"
+    name: "res4b7_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2b"
+    name: "res4b7_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2b"
+    name: "bn4b7_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2b"
+    name: "scale4b7_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b7_branch2b"
+    bottom: "res4b7_branch2b"
+    name: "res4b7_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2c"
+    name: "res4b7_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2c"
+    top: "res4b7_branch2c"
+    name: "bn4b7_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2c"
+    top: "res4b7_branch2c"
+    name: "scale4b7_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6"
+    bottom: "res4b7_branch2c"
+    top: "res4b7"
+    name: "res4b7"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b7"
+    top: "res4b7"
+    name: "res4b7_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7"
+    top: "res4b8_branch2a"
+    name: "res4b8_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2a"
+    name: "bn4b8_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2a"
+    name: "scale4b8_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b8_branch2a"
+    bottom: "res4b8_branch2a"
+    name: "res4b8_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2b"
+    name: "res4b8_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2b"
+    name: "bn4b8_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2b"
+    name: "scale4b8_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b8_branch2b"
+    bottom: "res4b8_branch2b"
+    name: "res4b8_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2c"
+    name: "res4b8_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2c"
+    top: "res4b8_branch2c"
+    name: "bn4b8_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2c"
+    top: "res4b8_branch2c"
+    name: "scale4b8_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7"
+    bottom: "res4b8_branch2c"
+    top: "res4b8"
+    name: "res4b8"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b8"
+    top: "res4b8"
+    name: "res4b8_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8"
+    top: "res4b9_branch2a"
+    name: "res4b9_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2a"
+    name: "bn4b9_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2a"
+    name: "scale4b9_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b9_branch2a"
+    bottom: "res4b9_branch2a"
+    name: "res4b9_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2b"
+    name: "res4b9_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2b"
+    name: "bn4b9_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2b"
+    name: "scale4b9_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b9_branch2b"
+    bottom: "res4b9_branch2b"
+    name: "res4b9_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2c"
+    name: "res4b9_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2c"
+    top: "res4b9_branch2c"
+    name: "bn4b9_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2c"
+    top: "res4b9_branch2c"
+    name: "scale4b9_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8"
+    bottom: "res4b9_branch2c"
+    top: "res4b9"
+    name: "res4b9"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b9"
+    top: "res4b9"
+    name: "res4b9_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9"
+    top: "res4b10_branch2a"
+    name: "res4b10_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2a"
+    name: "bn4b10_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2a"
+    name: "scale4b10_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b10_branch2a"
+    bottom: "res4b10_branch2a"
+    name: "res4b10_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2b"
+    name: "res4b10_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2b"
+    name: "bn4b10_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2b"
+    name: "scale4b10_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b10_branch2b"
+    bottom: "res4b10_branch2b"
+    name: "res4b10_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2c"
+    name: "res4b10_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2c"
+    top: "res4b10_branch2c"
+    name: "bn4b10_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2c"
+    top: "res4b10_branch2c"
+    name: "scale4b10_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9"
+    bottom: "res4b10_branch2c"
+    top: "res4b10"
+    name: "res4b10"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b10"
+    top: "res4b10"
+    name: "res4b10_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10"
+    top: "res4b11_branch2a"
+    name: "res4b11_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2a"
+    name: "bn4b11_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2a"
+    name: "scale4b11_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b11_branch2a"
+    bottom: "res4b11_branch2a"
+    name: "res4b11_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2b"
+    name: "res4b11_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2b"
+    name: "bn4b11_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2b"
+    name: "scale4b11_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b11_branch2b"
+    bottom: "res4b11_branch2b"
+    name: "res4b11_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2c"
+    name: "res4b11_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2c"
+    top: "res4b11_branch2c"
+    name: "bn4b11_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2c"
+    top: "res4b11_branch2c"
+    name: "scale4b11_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10"
+    bottom: "res4b11_branch2c"
+    top: "res4b11"
+    name: "res4b11"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b11"
+    top: "res4b11"
+    name: "res4b11_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11"
+    top: "res4b12_branch2a"
+    name: "res4b12_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2a"
+    name: "bn4b12_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2a"
+    name: "scale4b12_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b12_branch2a"
+    bottom: "res4b12_branch2a"
+    name: "res4b12_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2b"
+    name: "res4b12_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2b"
+    name: "bn4b12_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2b"
+    name: "scale4b12_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b12_branch2b"
+    bottom: "res4b12_branch2b"
+    name: "res4b12_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2c"
+    name: "res4b12_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2c"
+    top: "res4b12_branch2c"
+    name: "bn4b12_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2c"
+    top: "res4b12_branch2c"
+    name: "scale4b12_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11"
+    bottom: "res4b12_branch2c"
+    top: "res4b12"
+    name: "res4b12"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b12"
+    top: "res4b12"
+    name: "res4b12_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12"
+    top: "res4b13_branch2a"
+    name: "res4b13_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2a"
+    name: "bn4b13_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2a"
+    name: "scale4b13_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b13_branch2a"
+    bottom: "res4b13_branch2a"
+    name: "res4b13_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2b"
+    name: "res4b13_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2b"
+    name: "bn4b13_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2b"
+    name: "scale4b13_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b13_branch2b"
+    bottom: "res4b13_branch2b"
+    name: "res4b13_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2c"
+    name: "res4b13_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2c"
+    top: "res4b13_branch2c"
+    name: "bn4b13_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2c"
+    top: "res4b13_branch2c"
+    name: "scale4b13_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12"
+    bottom: "res4b13_branch2c"
+    top: "res4b13"
+    name: "res4b13"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b13"
+    top: "res4b13"
+    name: "res4b13_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13"
+    top: "res4b14_branch2a"
+    name: "res4b14_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2a"
+    name: "bn4b14_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2a"
+    name: "scale4b14_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b14_branch2a"
+    bottom: "res4b14_branch2a"
+    name: "res4b14_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2b"
+    name: "res4b14_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2b"
+    name: "bn4b14_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2b"
+    name: "scale4b14_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b14_branch2b"
+    bottom: "res4b14_branch2b"
+    name: "res4b14_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2c"
+    name: "res4b14_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2c"
+    top: "res4b14_branch2c"
+    name: "bn4b14_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2c"
+    top: "res4b14_branch2c"
+    name: "scale4b14_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13"
+    bottom: "res4b14_branch2c"
+    top: "res4b14"
+    name: "res4b14"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b14"
+    top: "res4b14"
+    name: "res4b14_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14"
+    top: "res4b15_branch2a"
+    name: "res4b15_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2a"
+    name: "bn4b15_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2a"
+    name: "scale4b15_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b15_branch2a"
+    bottom: "res4b15_branch2a"
+    name: "res4b15_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2b"
+    name: "res4b15_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2b"
+    name: "bn4b15_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2b"
+    name: "scale4b15_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b15_branch2b"
+    bottom: "res4b15_branch2b"
+    name: "res4b15_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2c"
+    name: "res4b15_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2c"
+    top: "res4b15_branch2c"
+    name: "bn4b15_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2c"
+    top: "res4b15_branch2c"
+    name: "scale4b15_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14"
+    bottom: "res4b15_branch2c"
+    top: "res4b15"
+    name: "res4b15"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b15"
+    top: "res4b15"
+    name: "res4b15_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15"
+    top: "res4b16_branch2a"
+    name: "res4b16_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2a"
+    name: "bn4b16_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2a"
+    name: "scale4b16_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b16_branch2a"
+    bottom: "res4b16_branch2a"
+    name: "res4b16_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2b"
+    name: "res4b16_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2b"
+    name: "bn4b16_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2b"
+    name: "scale4b16_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b16_branch2b"
+    bottom: "res4b16_branch2b"
+    name: "res4b16_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2c"
+    name: "res4b16_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2c"
+    top: "res4b16_branch2c"
+    name: "bn4b16_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2c"
+    top: "res4b16_branch2c"
+    name: "scale4b16_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15"
+    bottom: "res4b16_branch2c"
+    top: "res4b16"
+    name: "res4b16"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b16"
+    top: "res4b16"
+    name: "res4b16_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16"
+    top: "res4b17_branch2a"
+    name: "res4b17_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2a"
+    name: "bn4b17_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2a"
+    name: "scale4b17_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b17_branch2a"
+    bottom: "res4b17_branch2a"
+    name: "res4b17_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2b"
+    name: "res4b17_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2b"
+    name: "bn4b17_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2b"
+    name: "scale4b17_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b17_branch2b"
+    bottom: "res4b17_branch2b"
+    name: "res4b17_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2c"
+    name: "res4b17_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2c"
+    top: "res4b17_branch2c"
+    name: "bn4b17_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2c"
+    top: "res4b17_branch2c"
+    name: "scale4b17_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16"
+    bottom: "res4b17_branch2c"
+    top: "res4b17"
+    name: "res4b17"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b17"
+    top: "res4b17"
+    name: "res4b17_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17"
+    top: "res4b18_branch2a"
+    name: "res4b18_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2a"
+    name: "bn4b18_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2a"
+    name: "scale4b18_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b18_branch2a"
+    bottom: "res4b18_branch2a"
+    name: "res4b18_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2b"
+    name: "res4b18_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2b"
+    name: "bn4b18_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2b"
+    name: "scale4b18_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b18_branch2b"
+    bottom: "res4b18_branch2b"
+    name: "res4b18_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2c"
+    name: "res4b18_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2c"
+    top: "res4b18_branch2c"
+    name: "bn4b18_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2c"
+    top: "res4b18_branch2c"
+    name: "scale4b18_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17"
+    bottom: "res4b18_branch2c"
+    top: "res4b18"
+    name: "res4b18"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b18"
+    top: "res4b18"
+    name: "res4b18_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18"
+    top: "res4b19_branch2a"
+    name: "res4b19_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2a"
+    name: "bn4b19_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2a"
+    name: "scale4b19_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b19_branch2a"
+    bottom: "res4b19_branch2a"
+    name: "res4b19_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2b"
+    name: "res4b19_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2b"
+    name: "bn4b19_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2b"
+    name: "scale4b19_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b19_branch2b"
+    bottom: "res4b19_branch2b"
+    name: "res4b19_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2c"
+    name: "res4b19_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2c"
+    top: "res4b19_branch2c"
+    name: "bn4b19_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2c"
+    top: "res4b19_branch2c"
+    name: "scale4b19_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18"
+    bottom: "res4b19_branch2c"
+    top: "res4b19"
+    name: "res4b19"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b19"
+    top: "res4b19"
+    name: "res4b19_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19"
+    top: "res4b20_branch2a"
+    name: "res4b20_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2a"
+    name: "bn4b20_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2a"
+    name: "scale4b20_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b20_branch2a"
+    bottom: "res4b20_branch2a"
+    name: "res4b20_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2b"
+    name: "res4b20_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2b"
+    name: "bn4b20_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2b"
+    name: "scale4b20_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b20_branch2b"
+    bottom: "res4b20_branch2b"
+    name: "res4b20_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2c"
+    name: "res4b20_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2c"
+    top: "res4b20_branch2c"
+    name: "bn4b20_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2c"
+    top: "res4b20_branch2c"
+    name: "scale4b20_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19"
+    bottom: "res4b20_branch2c"
+    top: "res4b20"
+    name: "res4b20"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b20"
+    top: "res4b20"
+    name: "res4b20_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20"
+    top: "res4b21_branch2a"
+    name: "res4b21_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2a"
+    name: "bn4b21_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2a"
+    name: "scale4b21_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b21_branch2a"
+    bottom: "res4b21_branch2a"
+    name: "res4b21_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2b"
+    name: "res4b21_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2b"
+    name: "bn4b21_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2b"
+    name: "scale4b21_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b21_branch2b"
+    bottom: "res4b21_branch2b"
+    name: "res4b21_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2c"
+    name: "res4b21_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2c"
+    top: "res4b21_branch2c"
+    name: "bn4b21_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2c"
+    top: "res4b21_branch2c"
+    name: "scale4b21_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20"
+    bottom: "res4b21_branch2c"
+    top: "res4b21"
+    name: "res4b21"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b21"
+    top: "res4b21"
+    name: "res4b21_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21"
+    top: "res4b22_branch2a"
+    name: "res4b22_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2a"
+    name: "bn4b22_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2a"
+    name: "scale4b22_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b22_branch2a"
+    bottom: "res4b22_branch2a"
+    name: "res4b22_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2b"
+    name: "res4b22_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2b"
+    name: "bn4b22_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2b"
+    name: "scale4b22_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b22_branch2b"
+    bottom: "res4b22_branch2b"
+    name: "res4b22_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2c"
+    name: "res4b22_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2c"
+    top: "res4b22_branch2c"
+    name: "bn4b22_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2c"
+    top: "res4b22_branch2c"
+    name: "scale4b22_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21"
+    bottom: "res4b22_branch2c"
+    top: "res4b22"
+    name: "res4b22"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res4b22"
+    name: "res4b22_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res5a_branch1"
+    name: "res5a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "bn5a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "scale5a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "bn5a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "scale5a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5a_branch2a"
+    bottom: "res5a_branch2a"
+    name: "res5a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "bn5a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "scale5a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5a_branch2b"
+    bottom: "res5a_branch2b"
+    name: "res5a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2c"
+    name: "res5a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "bn5a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "scale5a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    bottom: "res5a_branch2c"
+    top: "res5a"
+    name: "res5a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5a"
+    name: "res5a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "bn5b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "scale5b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5b_branch2a"
+    bottom: "res5b_branch2a"
+    name: "res5b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "bn5b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "scale5b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5b_branch2b"
+    bottom: "res5b_branch2b"
+    name: "res5b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2c"
+    name: "res5b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "bn5b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "scale5b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a"
+    bottom: "res5b_branch2c"
+    top: "res5b"
+    name: "res5b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5b"
+    name: "res5b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "bn5c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "scale5c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5c_branch2a"
+    bottom: "res5c_branch2a"
+    name: "res5c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "bn5c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "scale5c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5c_branch2b"
+    bottom: "res5c_branch2b"
+    name: "res5c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2c"
+    name: "res5c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "bn5c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "scale5c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b"
+    bottom: "res5c_branch2c"
+    top: "res5c"
+    name: "res5c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5c"
+    top: "res5c"
+    name: "res5c_relu"
+    type: "ReLU"
+}
+
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "res5c"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+#  top: 'rpn_scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#layer {
+#  name: 'debug-data'
+#  type: 'Python'
+#  bottom: 'data'
+#  bottom: 'rpn_rois'
+#  bottom: 'rpn_scores'
+#  python_param {
+#    module: 'rpn.debug_layer'
+#    layer: 'RPNDebugLayer'
+#  }
+#}
+
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 2"
+  }
+}
+
+#----------------------new conv layer------------------
+layer {
+    bottom: "res5c"
+    top: "conv_new_1"
+    name: "conv_new_1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "conv_new_1"
+    name: "conv_new_1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_cls"
+    name: "rfcn_cls"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1029 #21*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_bbox"
+    name: "rfcn_bbox"
+    type: "Convolution"
+    convolution_param {
+        num_output: 392 #8*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+#--------------position sensitive RoI pooling--------------
+layer {
+    bottom: "rfcn_cls"
+    bottom: "rois"
+    top: "psroipooled_cls_rois"
+    name: "psroipooled_cls_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 21
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_cls_rois"
+    top: "cls_score"
+    name: "ave_cls_score_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+layer {
+    bottom: "rfcn_bbox"
+    bottom: "rois"
+    top: "psroipooled_loc_rois"
+    name: "psroipooled_loc_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 8
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_loc_rois"
+    top: "bbox_pred"
+    name: "ave_bbox_pred_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+#-----------------------output------------------------
+layer {
+   name: "loss"
+   type: "SoftmaxWithLoss"
+   bottom: "cls_score"
+   bottom: "labels"
+   top: "loss_cls"
+   loss_weight: 1
+   propagate_down: true
+   propagate_down: false
+}
+
+layer {
+   name: "accuarcy"
+   type: "Accuracy"
+   bottom: "cls_score"
+   bottom: "labels"
+   top: "accuarcy"
+   #include: { phase: TEST }
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "loss_bbox"
+   type: "SmoothL1LossOHEM"
+   bottom: "bbox_pred"
+   bottom: "bbox_targets"
+   bottom: 'bbox_inside_weights'
+   top: "loss_bbox"
+   loss_weight: 1
+   loss_param {
+        normalization: PRE_FIXED
+        pre_fixed_normalizer: 128
+   }
+   propagate_down: true
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+    name: "silence"
+    type: "Silence"
+    bottom: "bbox_outside_weights"
+}
+
diff --git a/models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic_ohem.prototxt b/models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic_ohem.prototxt
new file mode 100644
index 0000000..2c43883
--- /dev/null
+++ b/models/pascal_voc/ResNet-101/rfcn_end2end/train_agonistic_ohem.prototxt
@@ -0,0 +1,7344 @@
+name: "ResNet-101"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+# ------------------------ conv1 -----------------------------
+layer {
+    bottom: "data"
+    top: "conv1"
+    name: "conv1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 7
+        pad: 3
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "bn_conv1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "scale_conv1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "conv1"
+    bottom: "conv1"
+    name: "conv1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv1"
+    top: "pool1"
+    name: "pool1"
+    type: "Pooling"
+    pooling_param {
+        kernel_size: 3
+        stride: 2
+        pool: MAX
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch1"
+    name: "res2a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "bn2a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "scale2a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "bn2a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "scale2a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2a_branch2a"
+    bottom: "res2a_branch2a"
+    name: "res2a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "bn2a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "scale2a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2a_branch2b"
+    bottom: "res2a_branch2b"
+    name: "res2a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2c"
+    name: "res2a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "bn2a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "scale2a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    bottom: "res2a_branch2c"
+    top: "res2a"
+    name: "res2a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2a"
+    name: "res2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "bn2b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "scale2b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2b_branch2a"
+    bottom: "res2b_branch2a"
+    name: "res2b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "bn2b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "scale2b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2b_branch2b"
+    bottom: "res2b_branch2b"
+    name: "res2b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2c"
+    name: "res2b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "bn2b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "scale2b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a"
+    bottom: "res2b_branch2c"
+    top: "res2b"
+    name: "res2b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2b"
+    name: "res2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "bn2c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "scale2c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2c_branch2a"
+    bottom: "res2c_branch2a"
+    name: "res2c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "bn2c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "scale2c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res2c_branch2b"
+    bottom: "res2c_branch2b"
+    name: "res2c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2c"
+    name: "res2c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "bn2c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "scale2c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b"
+    bottom: "res2c_branch2c"
+    top: "res2c"
+    name: "res2c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res2c"
+    name: "res2c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch1"
+    name: "res3a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "bn3a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "scale3a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "bn3a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "scale3a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3a_branch2a"
+    bottom: "res3a_branch2a"
+    name: "res3a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "bn3a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "scale3a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3a_branch2b"
+    bottom: "res3a_branch2b"
+    name: "res3a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2c"
+    name: "res3a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "bn3a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "scale3a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    bottom: "res3a_branch2c"
+    top: "res3a"
+    name: "res3a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3a"
+    name: "res3a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3b1_branch2a"
+    name: "res3b1_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2a"
+    name: "bn3b1_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2a"
+    name: "scale3b1_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b1_branch2a"
+    bottom: "res3b1_branch2a"
+    name: "res3b1_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1_branch2a"
+    top: "res3b1_branch2b"
+    name: "res3b1_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2b"
+    name: "bn3b1_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2b"
+    name: "scale3b1_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b1_branch2b"
+    bottom: "res3b1_branch2b"
+    name: "res3b1_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1_branch2b"
+    top: "res3b1_branch2c"
+    name: "res3b1_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2c"
+    top: "res3b1_branch2c"
+    name: "bn3b1_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1_branch2c"
+    top: "res3b1_branch2c"
+    name: "scale3b1_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a"
+    bottom: "res3b1_branch2c"
+    top: "res3b1"
+    name: "res3b1"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b1"
+    top: "res3b1"
+    name: "res3b1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b1"
+    top: "res3b2_branch2a"
+    name: "res3b2_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2a"
+    name: "bn3b2_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2a"
+    name: "scale3b2_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b2_branch2a"
+    bottom: "res3b2_branch2a"
+    name: "res3b2_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2_branch2a"
+    top: "res3b2_branch2b"
+    name: "res3b2_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2b"
+    name: "bn3b2_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2b"
+    name: "scale3b2_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b2_branch2b"
+    bottom: "res3b2_branch2b"
+    name: "res3b2_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2_branch2b"
+    top: "res3b2_branch2c"
+    name: "res3b2_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2c"
+    top: "res3b2_branch2c"
+    name: "bn3b2_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2_branch2c"
+    top: "res3b2_branch2c"
+    name: "scale3b2_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b1"
+    bottom: "res3b2_branch2c"
+    top: "res3b2"
+    name: "res3b2"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b2"
+    top: "res3b2"
+    name: "res3b2_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b2"
+    top: "res3b3_branch2a"
+    name: "res3b3_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2a"
+    name: "bn3b3_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2a"
+    name: "scale3b3_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b3_branch2a"
+    bottom: "res3b3_branch2a"
+    name: "res3b3_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3_branch2a"
+    top: "res3b3_branch2b"
+    name: "res3b3_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2b"
+    name: "bn3b3_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2b"
+    name: "scale3b3_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res3b3_branch2b"
+    bottom: "res3b3_branch2b"
+    name: "res3b3_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3_branch2b"
+    top: "res3b3_branch2c"
+    name: "res3b3_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2c"
+    top: "res3b3_branch2c"
+    name: "bn3b3_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3_branch2c"
+    top: "res3b3_branch2c"
+    name: "scale3b3_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b2"
+    bottom: "res3b3_branch2c"
+    top: "res3b3"
+    name: "res3b3"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res3b3"
+    name: "res3b3_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res4a_branch1"
+    name: "res4a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "bn4a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "scale4a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b3"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "bn4a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "scale4a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4a_branch2a"
+    bottom: "res4a_branch2a"
+    name: "res4a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "bn4a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "scale4a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4a_branch2b"
+    bottom: "res4a_branch2b"
+    name: "res4a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2c"
+    name: "res4a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "bn4a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "scale4a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    bottom: "res4a_branch2c"
+    top: "res4a"
+    name: "res4a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4a"
+    name: "res4a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4b1_branch2a"
+    name: "res4b1_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2a"
+    name: "bn4b1_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2a"
+    name: "scale4b1_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b1_branch2a"
+    bottom: "res4b1_branch2a"
+    name: "res4b1_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1_branch2a"
+    top: "res4b1_branch2b"
+    name: "res4b1_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2b"
+    name: "bn4b1_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2b"
+    name: "scale4b1_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b1_branch2b"
+    bottom: "res4b1_branch2b"
+    name: "res4b1_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1_branch2b"
+    top: "res4b1_branch2c"
+    name: "res4b1_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2c"
+    top: "res4b1_branch2c"
+    name: "bn4b1_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1_branch2c"
+    top: "res4b1_branch2c"
+    name: "scale4b1_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a"
+    bottom: "res4b1_branch2c"
+    top: "res4b1"
+    name: "res4b1"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b1"
+    top: "res4b1"
+    name: "res4b1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b1"
+    top: "res4b2_branch2a"
+    name: "res4b2_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2a"
+    name: "bn4b2_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2a"
+    name: "scale4b2_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b2_branch2a"
+    bottom: "res4b2_branch2a"
+    name: "res4b2_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2_branch2a"
+    top: "res4b2_branch2b"
+    name: "res4b2_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2b"
+    name: "bn4b2_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2b"
+    name: "scale4b2_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b2_branch2b"
+    bottom: "res4b2_branch2b"
+    name: "res4b2_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2_branch2b"
+    top: "res4b2_branch2c"
+    name: "res4b2_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2c"
+    top: "res4b2_branch2c"
+    name: "bn4b2_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2_branch2c"
+    top: "res4b2_branch2c"
+    name: "scale4b2_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b1"
+    bottom: "res4b2_branch2c"
+    top: "res4b2"
+    name: "res4b2"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b2"
+    top: "res4b2"
+    name: "res4b2_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b2"
+    top: "res4b3_branch2a"
+    name: "res4b3_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2a"
+    name: "bn4b3_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2a"
+    name: "scale4b3_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b3_branch2a"
+    bottom: "res4b3_branch2a"
+    name: "res4b3_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3_branch2a"
+    top: "res4b3_branch2b"
+    name: "res4b3_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2b"
+    name: "bn4b3_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2b"
+    name: "scale4b3_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b3_branch2b"
+    bottom: "res4b3_branch2b"
+    name: "res4b3_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3_branch2b"
+    top: "res4b3_branch2c"
+    name: "res4b3_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2c"
+    top: "res4b3_branch2c"
+    name: "bn4b3_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3_branch2c"
+    top: "res4b3_branch2c"
+    name: "scale4b3_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b2"
+    bottom: "res4b3_branch2c"
+    top: "res4b3"
+    name: "res4b3"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b3"
+    top: "res4b3"
+    name: "res4b3_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b3"
+    top: "res4b4_branch2a"
+    name: "res4b4_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2a"
+    name: "bn4b4_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2a"
+    name: "scale4b4_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b4_branch2a"
+    bottom: "res4b4_branch2a"
+    name: "res4b4_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4_branch2a"
+    top: "res4b4_branch2b"
+    name: "res4b4_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2b"
+    name: "bn4b4_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2b"
+    name: "scale4b4_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b4_branch2b"
+    bottom: "res4b4_branch2b"
+    name: "res4b4_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4_branch2b"
+    top: "res4b4_branch2c"
+    name: "res4b4_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2c"
+    top: "res4b4_branch2c"
+    name: "bn4b4_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4_branch2c"
+    top: "res4b4_branch2c"
+    name: "scale4b4_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b3"
+    bottom: "res4b4_branch2c"
+    top: "res4b4"
+    name: "res4b4"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b4"
+    top: "res4b4"
+    name: "res4b4_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b4"
+    top: "res4b5_branch2a"
+    name: "res4b5_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2a"
+    name: "bn4b5_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2a"
+    name: "scale4b5_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b5_branch2a"
+    bottom: "res4b5_branch2a"
+    name: "res4b5_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5_branch2a"
+    top: "res4b5_branch2b"
+    name: "res4b5_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2b"
+    name: "bn4b5_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2b"
+    name: "scale4b5_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b5_branch2b"
+    bottom: "res4b5_branch2b"
+    name: "res4b5_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5_branch2b"
+    top: "res4b5_branch2c"
+    name: "res4b5_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2c"
+    top: "res4b5_branch2c"
+    name: "bn4b5_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5_branch2c"
+    top: "res4b5_branch2c"
+    name: "scale4b5_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b4"
+    bottom: "res4b5_branch2c"
+    top: "res4b5"
+    name: "res4b5"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b5"
+    top: "res4b5"
+    name: "res4b5_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b5"
+    top: "res4b6_branch2a"
+    name: "res4b6_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2a"
+    name: "bn4b6_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2a"
+    name: "scale4b6_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b6_branch2a"
+    bottom: "res4b6_branch2a"
+    name: "res4b6_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6_branch2a"
+    top: "res4b6_branch2b"
+    name: "res4b6_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2b"
+    name: "bn4b6_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2b"
+    name: "scale4b6_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b6_branch2b"
+    bottom: "res4b6_branch2b"
+    name: "res4b6_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6_branch2b"
+    top: "res4b6_branch2c"
+    name: "res4b6_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2c"
+    top: "res4b6_branch2c"
+    name: "bn4b6_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6_branch2c"
+    top: "res4b6_branch2c"
+    name: "scale4b6_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b5"
+    bottom: "res4b6_branch2c"
+    top: "res4b6"
+    name: "res4b6"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b6"
+    top: "res4b6"
+    name: "res4b6_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b6"
+    top: "res4b7_branch2a"
+    name: "res4b7_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2a"
+    name: "bn4b7_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2a"
+    name: "scale4b7_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b7_branch2a"
+    bottom: "res4b7_branch2a"
+    name: "res4b7_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7_branch2a"
+    top: "res4b7_branch2b"
+    name: "res4b7_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2b"
+    name: "bn4b7_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2b"
+    name: "scale4b7_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b7_branch2b"
+    bottom: "res4b7_branch2b"
+    name: "res4b7_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7_branch2b"
+    top: "res4b7_branch2c"
+    name: "res4b7_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2c"
+    top: "res4b7_branch2c"
+    name: "bn4b7_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7_branch2c"
+    top: "res4b7_branch2c"
+    name: "scale4b7_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b6"
+    bottom: "res4b7_branch2c"
+    top: "res4b7"
+    name: "res4b7"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b7"
+    top: "res4b7"
+    name: "res4b7_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b7"
+    top: "res4b8_branch2a"
+    name: "res4b8_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2a"
+    name: "bn4b8_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2a"
+    name: "scale4b8_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b8_branch2a"
+    bottom: "res4b8_branch2a"
+    name: "res4b8_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8_branch2a"
+    top: "res4b8_branch2b"
+    name: "res4b8_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2b"
+    name: "bn4b8_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2b"
+    name: "scale4b8_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b8_branch2b"
+    bottom: "res4b8_branch2b"
+    name: "res4b8_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8_branch2b"
+    top: "res4b8_branch2c"
+    name: "res4b8_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2c"
+    top: "res4b8_branch2c"
+    name: "bn4b8_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8_branch2c"
+    top: "res4b8_branch2c"
+    name: "scale4b8_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b7"
+    bottom: "res4b8_branch2c"
+    top: "res4b8"
+    name: "res4b8"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b8"
+    top: "res4b8"
+    name: "res4b8_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b8"
+    top: "res4b9_branch2a"
+    name: "res4b9_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2a"
+    name: "bn4b9_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2a"
+    name: "scale4b9_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b9_branch2a"
+    bottom: "res4b9_branch2a"
+    name: "res4b9_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9_branch2a"
+    top: "res4b9_branch2b"
+    name: "res4b9_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2b"
+    name: "bn4b9_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2b"
+    name: "scale4b9_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b9_branch2b"
+    bottom: "res4b9_branch2b"
+    name: "res4b9_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9_branch2b"
+    top: "res4b9_branch2c"
+    name: "res4b9_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2c"
+    top: "res4b9_branch2c"
+    name: "bn4b9_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9_branch2c"
+    top: "res4b9_branch2c"
+    name: "scale4b9_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b8"
+    bottom: "res4b9_branch2c"
+    top: "res4b9"
+    name: "res4b9"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b9"
+    top: "res4b9"
+    name: "res4b9_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b9"
+    top: "res4b10_branch2a"
+    name: "res4b10_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2a"
+    name: "bn4b10_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2a"
+    name: "scale4b10_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b10_branch2a"
+    bottom: "res4b10_branch2a"
+    name: "res4b10_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10_branch2a"
+    top: "res4b10_branch2b"
+    name: "res4b10_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2b"
+    name: "bn4b10_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2b"
+    name: "scale4b10_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b10_branch2b"
+    bottom: "res4b10_branch2b"
+    name: "res4b10_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10_branch2b"
+    top: "res4b10_branch2c"
+    name: "res4b10_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2c"
+    top: "res4b10_branch2c"
+    name: "bn4b10_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10_branch2c"
+    top: "res4b10_branch2c"
+    name: "scale4b10_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b9"
+    bottom: "res4b10_branch2c"
+    top: "res4b10"
+    name: "res4b10"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b10"
+    top: "res4b10"
+    name: "res4b10_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b10"
+    top: "res4b11_branch2a"
+    name: "res4b11_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2a"
+    name: "bn4b11_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2a"
+    name: "scale4b11_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b11_branch2a"
+    bottom: "res4b11_branch2a"
+    name: "res4b11_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11_branch2a"
+    top: "res4b11_branch2b"
+    name: "res4b11_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2b"
+    name: "bn4b11_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2b"
+    name: "scale4b11_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b11_branch2b"
+    bottom: "res4b11_branch2b"
+    name: "res4b11_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11_branch2b"
+    top: "res4b11_branch2c"
+    name: "res4b11_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2c"
+    top: "res4b11_branch2c"
+    name: "bn4b11_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11_branch2c"
+    top: "res4b11_branch2c"
+    name: "scale4b11_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b10"
+    bottom: "res4b11_branch2c"
+    top: "res4b11"
+    name: "res4b11"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b11"
+    top: "res4b11"
+    name: "res4b11_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b11"
+    top: "res4b12_branch2a"
+    name: "res4b12_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2a"
+    name: "bn4b12_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2a"
+    name: "scale4b12_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b12_branch2a"
+    bottom: "res4b12_branch2a"
+    name: "res4b12_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12_branch2a"
+    top: "res4b12_branch2b"
+    name: "res4b12_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2b"
+    name: "bn4b12_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2b"
+    name: "scale4b12_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b12_branch2b"
+    bottom: "res4b12_branch2b"
+    name: "res4b12_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12_branch2b"
+    top: "res4b12_branch2c"
+    name: "res4b12_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2c"
+    top: "res4b12_branch2c"
+    name: "bn4b12_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12_branch2c"
+    top: "res4b12_branch2c"
+    name: "scale4b12_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b11"
+    bottom: "res4b12_branch2c"
+    top: "res4b12"
+    name: "res4b12"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b12"
+    top: "res4b12"
+    name: "res4b12_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b12"
+    top: "res4b13_branch2a"
+    name: "res4b13_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2a"
+    name: "bn4b13_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2a"
+    name: "scale4b13_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b13_branch2a"
+    bottom: "res4b13_branch2a"
+    name: "res4b13_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13_branch2a"
+    top: "res4b13_branch2b"
+    name: "res4b13_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2b"
+    name: "bn4b13_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2b"
+    name: "scale4b13_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b13_branch2b"
+    bottom: "res4b13_branch2b"
+    name: "res4b13_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13_branch2b"
+    top: "res4b13_branch2c"
+    name: "res4b13_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2c"
+    top: "res4b13_branch2c"
+    name: "bn4b13_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13_branch2c"
+    top: "res4b13_branch2c"
+    name: "scale4b13_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b12"
+    bottom: "res4b13_branch2c"
+    top: "res4b13"
+    name: "res4b13"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b13"
+    top: "res4b13"
+    name: "res4b13_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b13"
+    top: "res4b14_branch2a"
+    name: "res4b14_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2a"
+    name: "bn4b14_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2a"
+    name: "scale4b14_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b14_branch2a"
+    bottom: "res4b14_branch2a"
+    name: "res4b14_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14_branch2a"
+    top: "res4b14_branch2b"
+    name: "res4b14_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2b"
+    name: "bn4b14_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2b"
+    name: "scale4b14_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b14_branch2b"
+    bottom: "res4b14_branch2b"
+    name: "res4b14_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14_branch2b"
+    top: "res4b14_branch2c"
+    name: "res4b14_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2c"
+    top: "res4b14_branch2c"
+    name: "bn4b14_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14_branch2c"
+    top: "res4b14_branch2c"
+    name: "scale4b14_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b13"
+    bottom: "res4b14_branch2c"
+    top: "res4b14"
+    name: "res4b14"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b14"
+    top: "res4b14"
+    name: "res4b14_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b14"
+    top: "res4b15_branch2a"
+    name: "res4b15_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2a"
+    name: "bn4b15_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2a"
+    name: "scale4b15_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b15_branch2a"
+    bottom: "res4b15_branch2a"
+    name: "res4b15_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15_branch2a"
+    top: "res4b15_branch2b"
+    name: "res4b15_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2b"
+    name: "bn4b15_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2b"
+    name: "scale4b15_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b15_branch2b"
+    bottom: "res4b15_branch2b"
+    name: "res4b15_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15_branch2b"
+    top: "res4b15_branch2c"
+    name: "res4b15_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2c"
+    top: "res4b15_branch2c"
+    name: "bn4b15_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15_branch2c"
+    top: "res4b15_branch2c"
+    name: "scale4b15_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b14"
+    bottom: "res4b15_branch2c"
+    top: "res4b15"
+    name: "res4b15"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b15"
+    top: "res4b15"
+    name: "res4b15_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b15"
+    top: "res4b16_branch2a"
+    name: "res4b16_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2a"
+    name: "bn4b16_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2a"
+    name: "scale4b16_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b16_branch2a"
+    bottom: "res4b16_branch2a"
+    name: "res4b16_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16_branch2a"
+    top: "res4b16_branch2b"
+    name: "res4b16_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2b"
+    name: "bn4b16_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2b"
+    name: "scale4b16_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b16_branch2b"
+    bottom: "res4b16_branch2b"
+    name: "res4b16_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16_branch2b"
+    top: "res4b16_branch2c"
+    name: "res4b16_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2c"
+    top: "res4b16_branch2c"
+    name: "bn4b16_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16_branch2c"
+    top: "res4b16_branch2c"
+    name: "scale4b16_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b15"
+    bottom: "res4b16_branch2c"
+    top: "res4b16"
+    name: "res4b16"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b16"
+    top: "res4b16"
+    name: "res4b16_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b16"
+    top: "res4b17_branch2a"
+    name: "res4b17_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2a"
+    name: "bn4b17_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2a"
+    name: "scale4b17_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b17_branch2a"
+    bottom: "res4b17_branch2a"
+    name: "res4b17_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17_branch2a"
+    top: "res4b17_branch2b"
+    name: "res4b17_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2b"
+    name: "bn4b17_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2b"
+    name: "scale4b17_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b17_branch2b"
+    bottom: "res4b17_branch2b"
+    name: "res4b17_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17_branch2b"
+    top: "res4b17_branch2c"
+    name: "res4b17_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2c"
+    top: "res4b17_branch2c"
+    name: "bn4b17_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17_branch2c"
+    top: "res4b17_branch2c"
+    name: "scale4b17_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b16"
+    bottom: "res4b17_branch2c"
+    top: "res4b17"
+    name: "res4b17"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b17"
+    top: "res4b17"
+    name: "res4b17_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b17"
+    top: "res4b18_branch2a"
+    name: "res4b18_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2a"
+    name: "bn4b18_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2a"
+    name: "scale4b18_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b18_branch2a"
+    bottom: "res4b18_branch2a"
+    name: "res4b18_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18_branch2a"
+    top: "res4b18_branch2b"
+    name: "res4b18_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2b"
+    name: "bn4b18_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2b"
+    name: "scale4b18_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b18_branch2b"
+    bottom: "res4b18_branch2b"
+    name: "res4b18_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18_branch2b"
+    top: "res4b18_branch2c"
+    name: "res4b18_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2c"
+    top: "res4b18_branch2c"
+    name: "bn4b18_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18_branch2c"
+    top: "res4b18_branch2c"
+    name: "scale4b18_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b17"
+    bottom: "res4b18_branch2c"
+    top: "res4b18"
+    name: "res4b18"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b18"
+    top: "res4b18"
+    name: "res4b18_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b18"
+    top: "res4b19_branch2a"
+    name: "res4b19_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2a"
+    name: "bn4b19_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2a"
+    name: "scale4b19_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b19_branch2a"
+    bottom: "res4b19_branch2a"
+    name: "res4b19_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19_branch2a"
+    top: "res4b19_branch2b"
+    name: "res4b19_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2b"
+    name: "bn4b19_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2b"
+    name: "scale4b19_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b19_branch2b"
+    bottom: "res4b19_branch2b"
+    name: "res4b19_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19_branch2b"
+    top: "res4b19_branch2c"
+    name: "res4b19_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2c"
+    top: "res4b19_branch2c"
+    name: "bn4b19_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19_branch2c"
+    top: "res4b19_branch2c"
+    name: "scale4b19_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b18"
+    bottom: "res4b19_branch2c"
+    top: "res4b19"
+    name: "res4b19"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b19"
+    top: "res4b19"
+    name: "res4b19_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b19"
+    top: "res4b20_branch2a"
+    name: "res4b20_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2a"
+    name: "bn4b20_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2a"
+    name: "scale4b20_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b20_branch2a"
+    bottom: "res4b20_branch2a"
+    name: "res4b20_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20_branch2a"
+    top: "res4b20_branch2b"
+    name: "res4b20_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2b"
+    name: "bn4b20_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2b"
+    name: "scale4b20_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b20_branch2b"
+    bottom: "res4b20_branch2b"
+    name: "res4b20_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20_branch2b"
+    top: "res4b20_branch2c"
+    name: "res4b20_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2c"
+    top: "res4b20_branch2c"
+    name: "bn4b20_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20_branch2c"
+    top: "res4b20_branch2c"
+    name: "scale4b20_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b19"
+    bottom: "res4b20_branch2c"
+    top: "res4b20"
+    name: "res4b20"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b20"
+    top: "res4b20"
+    name: "res4b20_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b20"
+    top: "res4b21_branch2a"
+    name: "res4b21_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2a"
+    name: "bn4b21_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2a"
+    name: "scale4b21_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b21_branch2a"
+    bottom: "res4b21_branch2a"
+    name: "res4b21_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21_branch2a"
+    top: "res4b21_branch2b"
+    name: "res4b21_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2b"
+    name: "bn4b21_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2b"
+    name: "scale4b21_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b21_branch2b"
+    bottom: "res4b21_branch2b"
+    name: "res4b21_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21_branch2b"
+    top: "res4b21_branch2c"
+    name: "res4b21_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2c"
+    top: "res4b21_branch2c"
+    name: "bn4b21_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21_branch2c"
+    top: "res4b21_branch2c"
+    name: "scale4b21_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b20"
+    bottom: "res4b21_branch2c"
+    top: "res4b21"
+    name: "res4b21"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b21"
+    top: "res4b21"
+    name: "res4b21_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b21"
+    top: "res4b22_branch2a"
+    name: "res4b22_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2a"
+    name: "bn4b22_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2a"
+    name: "scale4b22_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b22_branch2a"
+    bottom: "res4b22_branch2a"
+    name: "res4b22_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22_branch2a"
+    top: "res4b22_branch2b"
+    name: "res4b22_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2b"
+    name: "bn4b22_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2b"
+    name: "scale4b22_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res4b22_branch2b"
+    bottom: "res4b22_branch2b"
+    name: "res4b22_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22_branch2b"
+    top: "res4b22_branch2c"
+    name: "res4b22_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2c"
+    top: "res4b22_branch2c"
+    name: "bn4b22_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22_branch2c"
+    top: "res4b22_branch2c"
+    name: "scale4b22_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b21"
+    bottom: "res4b22_branch2c"
+    top: "res4b22"
+    name: "res4b22"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res4b22"
+    name: "res4b22_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res5a_branch1"
+    name: "res5a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "bn5a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "scale5a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b22"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "bn5a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "scale5a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5a_branch2a"
+    bottom: "res5a_branch2a"
+    name: "res5a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "bn5a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "scale5a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5a_branch2b"
+    bottom: "res5a_branch2b"
+    name: "res5a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2c"
+    name: "res5a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "bn5a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "scale5a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    bottom: "res5a_branch2c"
+    top: "res5a"
+    name: "res5a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5a"
+    name: "res5a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "bn5b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "scale5b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5b_branch2a"
+    bottom: "res5b_branch2a"
+    name: "res5b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "bn5b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "scale5b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5b_branch2b"
+    bottom: "res5b_branch2b"
+    name: "res5b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2c"
+    name: "res5b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "bn5b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "scale5b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a"
+    bottom: "res5b_branch2c"
+    top: "res5b"
+    name: "res5b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5b"
+    name: "res5b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "bn5c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "scale5c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5c_branch2a"
+    bottom: "res5c_branch2a"
+    name: "res5c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "bn5c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "scale5c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    top: "res5c_branch2b"
+    bottom: "res5c_branch2b"
+    name: "res5c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2c"
+    name: "res5c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "bn5c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "scale5c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b"
+    bottom: "res5c_branch2c"
+    top: "res5c"
+    name: "res5c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5c"
+    top: "res5c"
+    name: "res5c_relu"
+    type: "ReLU"
+}
+
+
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "res5c"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+#  top: 'rpn_scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#layer {
+#  name: 'debug-data'
+#  type: 'Python'
+#  bottom: 'data'
+#  bottom: 'rpn_rois'
+#  bottom: 'rpn_scores'
+#  python_param {
+#    module: 'rpn.debug_layer'
+#    layer: 'RPNDebugLayer'
+#  }
+#}
+
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 2"
+  }
+}
+
+#----------------------new conv layer------------------
+layer {
+    bottom: "res5c"
+    top: "conv_new_1"
+    name: "conv_new_1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "conv_new_1"
+    name: "conv_new_1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_cls"
+    name: "rfcn_cls"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1029 #21*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_bbox"
+    name: "rfcn_bbox"
+    type: "Convolution"
+    convolution_param {
+        num_output: 392 #8*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+#--------------position sensitive RoI pooling--------------
+layer {
+    bottom: "rfcn_cls"
+    bottom: "rois"
+    top: "psroipooled_cls_rois"
+    name: "psroipooled_cls_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 21
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_cls_rois"
+    top: "cls_score"
+    name: "ave_cls_score_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+layer {
+    bottom: "rfcn_bbox"
+    bottom: "rois"
+    top: "psroipooled_loc_rois"
+    name: "psroipooled_loc_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 8
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_loc_rois"
+    top: "bbox_pred"
+    name: "ave_bbox_pred_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+#--------------online hard example mining--------------
+layer {
+   name: "per_roi_loss_cls"
+   type: "SoftmaxWithLossOHEM"
+   bottom: "cls_score"
+   bottom: "labels"
+   top: "temp_loss_cls"
+   top: "temp_prob_cls"
+   top: "per_roi_loss_cls"
+   loss_weight: 0
+   loss_weight: 0
+   loss_weight: 0
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "per_roi_loss_bbox"
+   type: "SmoothL1LossOHEM"
+   bottom: "bbox_pred"
+   bottom: "bbox_targets"
+   bottom: "bbox_inside_weights"
+   top: "temp_loss_bbox"
+   top: "per_roi_loss_bbox"
+   loss_weight: 0
+   loss_weight: 0
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "per_roi_loss"
+   type: "Eltwise"
+   bottom: "per_roi_loss_cls"
+   bottom: "per_roi_loss_bbox"
+   top: "per_roi_loss"
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   bottom: "rois"
+   bottom: "per_roi_loss"
+   bottom: "labels"
+   bottom: "bbox_inside_weights"
+   top: "labels_ohem"
+   top: "bbox_loss_weights_ohem"
+   name: "annotator_detector"
+   type: "BoxAnnotatorOHEM"
+   box_annotator_ohem_param {
+        roi_per_img: 128
+        ignore_label: -1
+   }
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "silence"
+   type: "Silence"
+   bottom: "bbox_outside_weights"
+   bottom: "temp_loss_cls"
+   bottom: "temp_prob_cls"
+   bottom: "temp_loss_bbox"
+}
+
+#-----------------------output------------------------
+layer {
+   name: "loss"
+   type: "SoftmaxWithLoss"
+   bottom: "cls_score"
+   bottom: "labels_ohem"
+   top: "loss_cls"
+   loss_weight: 1
+   loss_param {
+        ignore_label: -1
+   }
+   propagate_down: true
+   propagate_down: false
+}
+
+layer {
+   name: "accuarcy"
+   type: "Accuracy"
+   bottom: "cls_score"
+   bottom: "labels_ohem"
+   top: "accuarcy"
+   #include: { phase: TEST }
+   accuracy_param {
+        ignore_label: -1
+   }
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "loss_bbox"
+   type: "SmoothL1LossOHEM"
+   bottom: "bbox_pred"
+   bottom: "bbox_targets"
+   bottom: "bbox_loss_weights_ohem"
+   top: "loss_bbox"
+   loss_weight: 1
+   loss_param {
+        normalization: PRE_FIXED
+        pre_fixed_normalizer: 128
+   }
+   propagate_down: true
+   propagate_down: false
+   propagate_down: false
+}
+
diff --git a/models/pascal_voc/ResNet-50/rfcn_end2end/class-aware/test.prototxt b/models/pascal_voc/ResNet-50/rfcn_end2end/class-aware/test.prototxt
new file mode 100644
index 0000000..83b223c
--- /dev/null
+++ b/models/pascal_voc/ResNet-50/rfcn_end2end/class-aware/test.prototxt
@@ -0,0 +1,3787 @@
+name: "ResNet50"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+layer {
+    bottom: "data"
+    top: "conv1"
+    name: "conv1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 7
+        pad: 3
+        stride: 2
+    }
+    param {
+        lr_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+    }
+    
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "bn_conv1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "scale_conv1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "conv1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv1"
+    top: "pool1"
+    name: "pool1"
+    type: "Pooling"
+    pooling_param {
+        kernel_size: 3
+        stride: 2
+        pool: MAX
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch1"
+    name: "res2a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "bn2a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "scale2a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "bn2a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "scale2a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "bn2a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "scale2a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2c"
+    name: "res2a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "bn2a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "scale2a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    bottom: "res2a_branch2c"
+    top: "res2a"
+    name: "res2a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2a"
+    name: "res2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "bn2b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "scale2b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "bn2b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "scale2b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2c"
+    name: "res2b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "bn2b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "scale2b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a"
+    bottom: "res2b_branch2c"
+    top: "res2b"
+    name: "res2b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2b"
+    name: "res2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "bn2c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "scale2c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "bn2c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "scale2c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2c"
+    name: "res2c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "bn2c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "scale2c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b"
+    bottom: "res2c_branch2c"
+    top: "res2c"
+    name: "res2c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res2c"
+    name: "res2c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch1"
+    name: "res3a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "bn3a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "scale3a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "bn3a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "scale3a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "bn3a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "scale3a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2c"
+    name: "res3a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "bn3a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "scale3a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    bottom: "res3a_branch2c"
+    top: "res3a"
+    name: "res3a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3a"
+    name: "res3a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3b_branch2a"
+    name: "res3b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "bn3b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "scale3b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "res3b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2b"
+    name: "res3b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "bn3b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "scale3b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "res3b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2c"
+    name: "res3b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2c"
+    top: "res3b_branch2c"
+    name: "bn3b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2c"
+    top: "res3b_branch2c"
+    name: "scale3b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a"
+    bottom: "res3b_branch2c"
+    top: "res3b"
+    name: "res3b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b"
+    top: "res3b"
+    name: "res3b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b"
+    top: "res3c_branch2a"
+    name: "res3c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "bn3c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "scale3c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "res3c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2b"
+    name: "res3c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "bn3c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "scale3c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "res3c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2c"
+    name: "res3c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2c"
+    top: "res3c_branch2c"
+    name: "bn3c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2c"
+    top: "res3c_branch2c"
+    name: "scale3c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b"
+    bottom: "res3c_branch2c"
+    top: "res3c"
+    name: "res3c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3c"
+    top: "res3c"
+    name: "res3c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c"
+    top: "res3d_branch2a"
+    name: "res3d_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "bn3d_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "scale3d_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "res3d_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2b"
+    name: "res3d_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "bn3d_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "scale3d_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "res3d_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2c"
+    name: "res3d_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2c"
+    top: "res3d_branch2c"
+    name: "bn3d_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2c"
+    top: "res3d_branch2c"
+    name: "scale3d_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c"
+    bottom: "res3d_branch2c"
+    top: "res3d"
+    name: "res3d"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3d"
+    top: "res3d"
+    name: "res3d_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d"
+    top: "res4a_branch1"
+    name: "res4a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "bn4a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "scale4a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "bn4a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "scale4a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "bn4a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "scale4a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2c"
+    name: "res4a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "bn4a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "scale4a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    bottom: "res4a_branch2c"
+    top: "res4a"
+    name: "res4a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4a"
+    name: "res4a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4b_branch2a"
+    name: "res4b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "bn4b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "scale4b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "res4b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2b"
+    name: "res4b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "bn4b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "scale4b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "res4b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2c"
+    name: "res4b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2c"
+    top: "res4b_branch2c"
+    name: "bn4b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2c"
+    top: "res4b_branch2c"
+    name: "scale4b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a"
+    bottom: "res4b_branch2c"
+    top: "res4b"
+    name: "res4b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b"
+    top: "res4b"
+    name: "res4b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b"
+    top: "res4c_branch2a"
+    name: "res4c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "bn4c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "scale4c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "res4c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2b"
+    name: "res4c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "bn4c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "scale4c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "res4c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2c"
+    name: "res4c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2c"
+    top: "res4c_branch2c"
+    name: "bn4c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2c"
+    top: "res4c_branch2c"
+    name: "scale4c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b"
+    bottom: "res4c_branch2c"
+    top: "res4c"
+    name: "res4c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4c"
+    top: "res4c"
+    name: "res4c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c"
+    top: "res4d_branch2a"
+    name: "res4d_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "bn4d_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "scale4d_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "res4d_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2b"
+    name: "res4d_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "bn4d_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "scale4d_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "res4d_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2c"
+    name: "res4d_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2c"
+    top: "res4d_branch2c"
+    name: "bn4d_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2c"
+    top: "res4d_branch2c"
+    name: "scale4d_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c"
+    bottom: "res4d_branch2c"
+    top: "res4d"
+    name: "res4d"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4d"
+    top: "res4d"
+    name: "res4d_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d"
+    top: "res4e_branch2a"
+    name: "res4e_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "bn4e_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "scale4e_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "res4e_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2b"
+    name: "res4e_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "bn4e_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "scale4e_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "res4e_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2c"
+    name: "res4e_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2c"
+    top: "res4e_branch2c"
+    name: "bn4e_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2c"
+    top: "res4e_branch2c"
+    name: "scale4e_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d"
+    bottom: "res4e_branch2c"
+    top: "res4e"
+    name: "res4e"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4e"
+    top: "res4e"
+    name: "res4e_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e"
+    top: "res4f_branch2a"
+    name: "res4f_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "bn4f_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "scale4f_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "res4f_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2b"
+    name: "res4f_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "bn4f_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "scale4f_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "res4f_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2c"
+    name: "res4f_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2c"
+    top: "res4f_branch2c"
+    name: "bn4f_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2c"
+    top: "res4f_branch2c"
+    name: "scale4f_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e"
+    bottom: "res4f_branch2c"
+    top: "res4f"
+    name: "res4f"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4f"
+    top: "res4f"
+    name: "res4f_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f"
+    top: "res5a_branch1"
+    name: "res5a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "bn5a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "scale5a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "bn5a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "scale5a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "bn5a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "scale5a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2c"
+    name: "res5a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "bn5a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "scale5a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    bottom: "res5a_branch2c"
+    top: "res5a"
+    name: "res5a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5a"
+    name: "res5a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "bn5b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "scale5b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "bn5b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "scale5b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2c"
+    name: "res5b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "bn5b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "scale5b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a"
+    bottom: "res5b_branch2c"
+    top: "res5b"
+    name: "res5b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5b"
+    name: "res5b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "bn5c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "scale5c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "bn5c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "scale5c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2c"
+    name: "res5c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "bn5c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "scale5c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b"
+    bottom: "res5c_branch2c"
+    top: "res5c"
+    name: "res5c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5c"
+    top: "res5c"
+    name: "res5c_relu"
+    type: "ReLU"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "res5c"
+  top: "rpn/output"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#----------------------new conv layer------------------
+layer {
+    bottom: "res5c"
+    top: "conv_new_1"
+    name: "conv_new_1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "conv_new_1"
+    name: "conv_new_1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_cls"
+    name: "rfcn_cls"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1029 #21*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_bbox"
+    name: "rfcn_bbox"
+    type: "Convolution"
+    convolution_param {
+        num_output: 4116 #8*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+#--------------position sensitive RoI pooling--------------
+layer {
+    bottom: "rfcn_cls"
+    bottom: "rois"
+    top: "psroipooled_cls_rois"
+    name: "psroipooled_cls_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 21
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_cls_rois"
+    top: "cls_score"
+    name: "ave_cls_score_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+layer {
+    bottom: "rfcn_bbox"
+    bottom: "rois"
+    top: "psroipooled_loc_rois"
+    name: "psroipooled_loc_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 84
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_loc_rois"
+    top: "bbox_pred_pre"
+    name: "ave_bbox_pred_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+#-----------------------output------------------------
+layer {
+    name: "cls_prob"
+    type: "Softmax"
+    bottom: "cls_score"
+    top: "cls_prob_pre"
+}
+
+layer {
+    name: "cls_prob_reshape"
+    type: "Reshape"
+    bottom: "cls_prob_pre"
+    top: "cls_prob"
+    reshape_param {
+        shape {
+            dim: -1
+            dim: 21
+        }
+    }
+}
+
+layer {
+    name: "bbox_pred_reshape"
+    type: "Reshape"
+    bottom: "bbox_pred_pre"
+    top: "bbox_pred"
+    reshape_param {
+        shape {
+            dim: -1
+            dim: 84
+        }
+    }
+}
+
+
diff --git a/models/pascal_voc/ResNet-50/rfcn_end2end/class-aware/train_ohem.prototxt b/models/pascal_voc/ResNet-50/rfcn_end2end/class-aware/train_ohem.prototxt
new file mode 100644
index 0000000..f709103
--- /dev/null
+++ b/models/pascal_voc/ResNet-50/rfcn_end2end/class-aware/train_ohem.prototxt
@@ -0,0 +1,3946 @@
+name: "ResNet-50"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+# ------------------------ conv1 -----------------------------
+layer {
+    bottom: "data"
+    top: "conv1"
+    name: "conv1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 7
+        pad: 3
+        stride: 2
+    }
+    param {
+        lr_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+    }
+    
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "bn_conv1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "scale_conv1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "conv1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv1"
+    top: "pool1"
+    name: "pool1"
+    type: "Pooling"
+    pooling_param {
+        kernel_size: 3
+        stride: 2
+        pool: MAX
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch1"
+    name: "res2a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "bn2a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "scale2a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "bn2a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "scale2a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "bn2a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "scale2a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2c"
+    name: "res2a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "bn2a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "scale2a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    bottom: "res2a_branch2c"
+    top: "res2a"
+    name: "res2a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2a"
+    name: "res2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "bn2b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "scale2b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "bn2b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "scale2b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2c"
+    name: "res2b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "bn2b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "scale2b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a"
+    bottom: "res2b_branch2c"
+    top: "res2b"
+    name: "res2b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2b"
+    name: "res2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "bn2c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "scale2c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "bn2c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "scale2c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2c"
+    name: "res2c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "bn2c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "scale2c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b"
+    bottom: "res2c_branch2c"
+    top: "res2c"
+    name: "res2c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res2c"
+    name: "res2c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch1"
+    name: "res3a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "bn3a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "scale3a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "bn3a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "scale3a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "bn3a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "scale3a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2c"
+    name: "res3a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "bn3a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "scale3a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    bottom: "res3a_branch2c"
+    top: "res3a"
+    name: "res3a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3a"
+    name: "res3a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3b_branch2a"
+    name: "res3b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "bn3b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "scale3b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "res3b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2b"
+    name: "res3b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "bn3b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "scale3b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "res3b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2c"
+    name: "res3b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2c"
+    top: "res3b_branch2c"
+    name: "bn3b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2c"
+    top: "res3b_branch2c"
+    name: "scale3b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a"
+    bottom: "res3b_branch2c"
+    top: "res3b"
+    name: "res3b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b"
+    top: "res3b"
+    name: "res3b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b"
+    top: "res3c_branch2a"
+    name: "res3c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "bn3c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "scale3c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "res3c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2b"
+    name: "res3c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "bn3c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "scale3c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "res3c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2c"
+    name: "res3c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2c"
+    top: "res3c_branch2c"
+    name: "bn3c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2c"
+    top: "res3c_branch2c"
+    name: "scale3c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b"
+    bottom: "res3c_branch2c"
+    top: "res3c"
+    name: "res3c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3c"
+    top: "res3c"
+    name: "res3c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c"
+    top: "res3d_branch2a"
+    name: "res3d_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "bn3d_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "scale3d_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "res3d_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2b"
+    name: "res3d_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "bn3d_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "scale3d_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "res3d_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2c"
+    name: "res3d_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2c"
+    top: "res3d_branch2c"
+    name: "bn3d_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2c"
+    top: "res3d_branch2c"
+    name: "scale3d_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c"
+    bottom: "res3d_branch2c"
+    top: "res3d"
+    name: "res3d"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3d"
+    top: "res3d"
+    name: "res3d_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d"
+    top: "res4a_branch1"
+    name: "res4a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "bn4a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "scale4a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "bn4a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "scale4a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "bn4a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "scale4a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2c"
+    name: "res4a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "bn4a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "scale4a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    bottom: "res4a_branch2c"
+    top: "res4a"
+    name: "res4a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4a"
+    name: "res4a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4b_branch2a"
+    name: "res4b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "bn4b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "scale4b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "res4b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2b"
+    name: "res4b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "bn4b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "scale4b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "res4b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2c"
+    name: "res4b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2c"
+    top: "res4b_branch2c"
+    name: "bn4b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2c"
+    top: "res4b_branch2c"
+    name: "scale4b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a"
+    bottom: "res4b_branch2c"
+    top: "res4b"
+    name: "res4b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b"
+    top: "res4b"
+    name: "res4b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b"
+    top: "res4c_branch2a"
+    name: "res4c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "bn4c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "scale4c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "res4c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2b"
+    name: "res4c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "bn4c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "scale4c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "res4c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2c"
+    name: "res4c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2c"
+    top: "res4c_branch2c"
+    name: "bn4c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2c"
+    top: "res4c_branch2c"
+    name: "scale4c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b"
+    bottom: "res4c_branch2c"
+    top: "res4c"
+    name: "res4c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4c"
+    top: "res4c"
+    name: "res4c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c"
+    top: "res4d_branch2a"
+    name: "res4d_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "bn4d_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "scale4d_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "res4d_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2b"
+    name: "res4d_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "bn4d_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "scale4d_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "res4d_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2c"
+    name: "res4d_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2c"
+    top: "res4d_branch2c"
+    name: "bn4d_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2c"
+    top: "res4d_branch2c"
+    name: "scale4d_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c"
+    bottom: "res4d_branch2c"
+    top: "res4d"
+    name: "res4d"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4d"
+    top: "res4d"
+    name: "res4d_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d"
+    top: "res4e_branch2a"
+    name: "res4e_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "bn4e_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "scale4e_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "res4e_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2b"
+    name: "res4e_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "bn4e_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "scale4e_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "res4e_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2c"
+    name: "res4e_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2c"
+    top: "res4e_branch2c"
+    name: "bn4e_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2c"
+    top: "res4e_branch2c"
+    name: "scale4e_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d"
+    bottom: "res4e_branch2c"
+    top: "res4e"
+    name: "res4e"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4e"
+    top: "res4e"
+    name: "res4e_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e"
+    top: "res4f_branch2a"
+    name: "res4f_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "bn4f_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "scale4f_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "res4f_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2b"
+    name: "res4f_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "bn4f_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "scale4f_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "res4f_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2c"
+    name: "res4f_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2c"
+    top: "res4f_branch2c"
+    name: "bn4f_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2c"
+    top: "res4f_branch2c"
+    name: "scale4f_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e"
+    bottom: "res4f_branch2c"
+    top: "res4f"
+    name: "res4f"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4f"
+    top: "res4f"
+    name: "res4f_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f"
+    top: "res5a_branch1"
+    name: "res5a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "bn5a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "scale5a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "bn5a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "scale5a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "bn5a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "scale5a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2c"
+    name: "res5a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "bn5a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "scale5a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    bottom: "res5a_branch2c"
+    top: "res5a"
+    name: "res5a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5a"
+    name: "res5a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "bn5b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "scale5b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "bn5b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "scale5b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2c"
+    name: "res5b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "bn5b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "scale5b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a"
+    bottom: "res5b_branch2c"
+    top: "res5b"
+    name: "res5b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5b"
+    name: "res5b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "bn5c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "scale5c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "bn5c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "scale5c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2c"
+    name: "res5c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "bn5c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "scale5c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b"
+    bottom: "res5c_branch2c"
+    top: "res5c"
+    name: "res5c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5c"
+    top: "res5c"
+    name: "res5c_relu"
+    type: "ReLU"
+}
+
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "res5c"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+#  top: 'rpn_scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#layer {
+#  name: 'debug-data'
+#  type: 'Python'
+#  bottom: 'data'
+#  bottom: 'rpn_rois'
+#  bottom: 'rpn_scores'
+#  python_param {
+#    module: 'rpn.debug_layer'
+#    layer: 'RPNDebugLayer'
+#  }
+#}
+
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#----------------------new conv layer------------------
+layer {
+    bottom: "res5c"
+    top: "conv_new_1"
+    name: "conv_new_1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "conv_new_1"
+    name: "conv_new_1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_cls"
+    name: "rfcn_cls"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1029 #21*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_bbox"
+    name: "rfcn_bbox"
+    type: "Convolution"
+    convolution_param {
+        num_output: 4116 #8*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+#--------------position sensitive RoI pooling--------------
+layer {
+    bottom: "rfcn_cls"
+    bottom: "rois"
+    top: "psroipooled_cls_rois"
+    name: "psroipooled_cls_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 21
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_cls_rois"
+    top: "cls_score"
+    name: "ave_cls_score_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+layer {
+    bottom: "rfcn_bbox"
+    bottom: "rois"
+    top: "psroipooled_loc_rois"
+    name: "psroipooled_loc_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 84
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_loc_rois"
+    top: "bbox_pred"
+    name: "ave_bbox_pred_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+#--------------online hard example mining--------------
+layer {
+   name: "per_roi_loss_cls"
+   type: "SoftmaxWithLossOHEM"
+   bottom: "cls_score"
+   bottom: "labels"
+   top: "temp_loss_cls"
+   top: "temp_prob_cls"
+   top: "per_roi_loss_cls"
+   loss_weight: 0
+   loss_weight: 0
+   loss_weight: 0
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "per_roi_loss_bbox"
+   type: "SmoothL1LossOHEM"
+   bottom: "bbox_pred"
+   bottom: "bbox_targets"
+   bottom: "bbox_inside_weights"
+   top: "temp_loss_bbox"
+   top: "per_roi_loss_bbox"
+   loss_weight: 0
+   loss_weight: 0
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "per_roi_loss"
+   type: "Eltwise"
+   bottom: "per_roi_loss_cls"
+   bottom: "per_roi_loss_bbox"
+   top: "per_roi_loss"
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   bottom: "rois"
+   bottom: "per_roi_loss"
+   bottom: "labels"
+   bottom: "bbox_inside_weights"
+   top: "labels_ohem"
+   top: "bbox_loss_weights_ohem"
+   name: "annotator_detector"
+   type: "BoxAnnotatorOHEM"
+   box_annotator_ohem_param {
+        roi_per_img: 128
+        ignore_label: -1
+   }
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "silence"
+   type: "Silence"
+   bottom: "bbox_outside_weights"
+   bottom: "temp_loss_cls"
+   bottom: "temp_prob_cls"
+   bottom: "temp_loss_bbox"
+}
+
+#-----------------------output------------------------
+layer {
+   name: "loss"
+   type: "SoftmaxWithLoss"
+   bottom: "cls_score"
+   bottom: "labels_ohem"
+   top: "loss_cls"
+   loss_weight: 1
+   loss_param {
+        ignore_label: -1
+   }
+   propagate_down: true
+   propagate_down: false
+}
+
+layer {
+   name: "accuarcy"
+   type: "Accuracy"
+   bottom: "cls_score"
+   bottom: "labels_ohem"
+   top: "accuarcy"
+   #include: { phase: TEST }
+   accuracy_param {
+        ignore_label: -1
+   }
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "loss_bbox"
+   type: "SmoothL1LossOHEM"
+   bottom: "bbox_pred"
+   bottom: "bbox_targets"
+   bottom: "bbox_loss_weights_ohem"
+   top: "loss_bbox"
+   loss_weight: 1
+   loss_param {
+        normalization: PRE_FIXED
+        pre_fixed_normalizer: 128
+   }
+   propagate_down: true
+   propagate_down: false
+   propagate_down: false
+}
+
diff --git a/models/pascal_voc/ResNet-50/rfcn_end2end/solver.prototxt b/models/pascal_voc/ResNet-50/rfcn_end2end/solver.prototxt
new file mode 100644
index 0000000..873a4bc
--- /dev/null
+++ b/models/pascal_voc/ResNet-50/rfcn_end2end/solver.prototxt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ResNet-50/rfcn_end2end/train_agonistic.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 80000
+display: 20
+
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "resnet50_rfcn"
+iter_size: 2
+# debug_info: true
diff --git a/models/pascal_voc/ResNet-50/rfcn_end2end/solver_ohem.prototxt b/models/pascal_voc/ResNet-50/rfcn_end2end/solver_ohem.prototxt
new file mode 100644
index 0000000..c8d196b
--- /dev/null
+++ b/models/pascal_voc/ResNet-50/rfcn_end2end/solver_ohem.prototxt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ResNet-50/rfcn_end2end/train_agonistic_ohem.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 80000
+display: 20
+
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "resnet50_rfcn_ohem"
+iter_size: 2
+# debug_info: true
diff --git a/models/pascal_voc/ResNet-50/rfcn_end2end/solver_warmup.prototxt b/models/pascal_voc/ResNet-50/rfcn_end2end/solver_warmup.prototxt
new file mode 100644
index 0000000..42274a9
--- /dev/null
+++ b/models/pascal_voc/ResNet-50/rfcn_end2end/solver_warmup.prototxt
@@ -0,0 +1,14 @@
+train_net: "models/pascal_voc/ResNet-50/rfcn_end2end/train_agonistic_ohem.prototxt"
+lr_policy: "fixed"
+base_lr: 0.0001
+
+display: 20
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "resnet50_rfcn_ohem"
+iter_size: 2
+# debug_info: true
diff --git a/models/pascal_voc/ResNet-50/rfcn_end2end/solver_warmup_continue.prototxt b/models/pascal_voc/ResNet-50/rfcn_end2end/solver_warmup_continue.prototxt
new file mode 100644
index 0000000..07255ca
--- /dev/null
+++ b/models/pascal_voc/ResNet-50/rfcn_end2end/solver_warmup_continue.prototxt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ResNet-50/rfcn_end2end/train_agonistic_ohem.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 70000
+
+display: 20
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "resnet50_rfcn_ohem"
+iter_size: 2
+# debug_info: true
diff --git a/models/pascal_voc/ResNet-50/rfcn_end2end/test_agonistic.prototxt b/models/pascal_voc/ResNet-50/rfcn_end2end/test_agonistic.prototxt
new file mode 100644
index 0000000..898e8eb
--- /dev/null
+++ b/models/pascal_voc/ResNet-50/rfcn_end2end/test_agonistic.prototxt
@@ -0,0 +1,3787 @@
+name: "ResNet50"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+layer {
+    bottom: "data"
+    top: "conv1"
+    name: "conv1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 7
+        pad: 3
+        stride: 2
+    }
+    param {
+        lr_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+    }
+    
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "bn_conv1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "scale_conv1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "conv1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv1"
+    top: "pool1"
+    name: "pool1"
+    type: "Pooling"
+    pooling_param {
+        kernel_size: 3
+        stride: 2
+        pool: MAX
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch1"
+    name: "res2a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "bn2a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "scale2a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "bn2a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "scale2a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "bn2a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "scale2a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2c"
+    name: "res2a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "bn2a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "scale2a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    bottom: "res2a_branch2c"
+    top: "res2a"
+    name: "res2a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2a"
+    name: "res2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "bn2b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "scale2b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "bn2b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "scale2b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2c"
+    name: "res2b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "bn2b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "scale2b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a"
+    bottom: "res2b_branch2c"
+    top: "res2b"
+    name: "res2b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2b"
+    name: "res2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "bn2c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "scale2c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "bn2c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "scale2c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2c"
+    name: "res2c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "bn2c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "scale2c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b"
+    bottom: "res2c_branch2c"
+    top: "res2c"
+    name: "res2c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res2c"
+    name: "res2c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch1"
+    name: "res3a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "bn3a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "scale3a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "bn3a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "scale3a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "bn3a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "scale3a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2c"
+    name: "res3a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "bn3a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "scale3a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    bottom: "res3a_branch2c"
+    top: "res3a"
+    name: "res3a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3a"
+    name: "res3a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3b_branch2a"
+    name: "res3b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "bn3b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "scale3b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "res3b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2b"
+    name: "res3b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "bn3b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "scale3b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "res3b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2c"
+    name: "res3b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2c"
+    top: "res3b_branch2c"
+    name: "bn3b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2c"
+    top: "res3b_branch2c"
+    name: "scale3b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a"
+    bottom: "res3b_branch2c"
+    top: "res3b"
+    name: "res3b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b"
+    top: "res3b"
+    name: "res3b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b"
+    top: "res3c_branch2a"
+    name: "res3c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "bn3c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "scale3c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "res3c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2b"
+    name: "res3c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "bn3c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "scale3c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "res3c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2c"
+    name: "res3c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2c"
+    top: "res3c_branch2c"
+    name: "bn3c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2c"
+    top: "res3c_branch2c"
+    name: "scale3c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b"
+    bottom: "res3c_branch2c"
+    top: "res3c"
+    name: "res3c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3c"
+    top: "res3c"
+    name: "res3c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c"
+    top: "res3d_branch2a"
+    name: "res3d_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "bn3d_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "scale3d_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "res3d_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2b"
+    name: "res3d_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "bn3d_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "scale3d_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "res3d_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2c"
+    name: "res3d_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2c"
+    top: "res3d_branch2c"
+    name: "bn3d_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2c"
+    top: "res3d_branch2c"
+    name: "scale3d_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c"
+    bottom: "res3d_branch2c"
+    top: "res3d"
+    name: "res3d"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3d"
+    top: "res3d"
+    name: "res3d_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d"
+    top: "res4a_branch1"
+    name: "res4a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "bn4a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "scale4a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "bn4a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "scale4a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "bn4a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "scale4a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2c"
+    name: "res4a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "bn4a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "scale4a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    bottom: "res4a_branch2c"
+    top: "res4a"
+    name: "res4a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4a"
+    name: "res4a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4b_branch2a"
+    name: "res4b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "bn4b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "scale4b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "res4b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2b"
+    name: "res4b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "bn4b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "scale4b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "res4b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2c"
+    name: "res4b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2c"
+    top: "res4b_branch2c"
+    name: "bn4b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2c"
+    top: "res4b_branch2c"
+    name: "scale4b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a"
+    bottom: "res4b_branch2c"
+    top: "res4b"
+    name: "res4b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b"
+    top: "res4b"
+    name: "res4b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b"
+    top: "res4c_branch2a"
+    name: "res4c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "bn4c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "scale4c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "res4c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2b"
+    name: "res4c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "bn4c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "scale4c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "res4c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2c"
+    name: "res4c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2c"
+    top: "res4c_branch2c"
+    name: "bn4c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2c"
+    top: "res4c_branch2c"
+    name: "scale4c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b"
+    bottom: "res4c_branch2c"
+    top: "res4c"
+    name: "res4c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4c"
+    top: "res4c"
+    name: "res4c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c"
+    top: "res4d_branch2a"
+    name: "res4d_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "bn4d_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "scale4d_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "res4d_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2b"
+    name: "res4d_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "bn4d_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "scale4d_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "res4d_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2c"
+    name: "res4d_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2c"
+    top: "res4d_branch2c"
+    name: "bn4d_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2c"
+    top: "res4d_branch2c"
+    name: "scale4d_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c"
+    bottom: "res4d_branch2c"
+    top: "res4d"
+    name: "res4d"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4d"
+    top: "res4d"
+    name: "res4d_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d"
+    top: "res4e_branch2a"
+    name: "res4e_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "bn4e_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "scale4e_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "res4e_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2b"
+    name: "res4e_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "bn4e_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "scale4e_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "res4e_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2c"
+    name: "res4e_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2c"
+    top: "res4e_branch2c"
+    name: "bn4e_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2c"
+    top: "res4e_branch2c"
+    name: "scale4e_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d"
+    bottom: "res4e_branch2c"
+    top: "res4e"
+    name: "res4e"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4e"
+    top: "res4e"
+    name: "res4e_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e"
+    top: "res4f_branch2a"
+    name: "res4f_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "bn4f_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "scale4f_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "res4f_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2b"
+    name: "res4f_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "bn4f_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "scale4f_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "res4f_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2c"
+    name: "res4f_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2c"
+    top: "res4f_branch2c"
+    name: "bn4f_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2c"
+    top: "res4f_branch2c"
+    name: "scale4f_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e"
+    bottom: "res4f_branch2c"
+    top: "res4f"
+    name: "res4f"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4f"
+    top: "res4f"
+    name: "res4f_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f"
+    top: "res5a_branch1"
+    name: "res5a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "bn5a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "scale5a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "bn5a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "scale5a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "bn5a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "scale5a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2c"
+    name: "res5a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "bn5a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "scale5a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    bottom: "res5a_branch2c"
+    top: "res5a"
+    name: "res5a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5a"
+    name: "res5a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "bn5b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "scale5b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "bn5b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "scale5b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2c"
+    name: "res5b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "bn5b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "scale5b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a"
+    bottom: "res5b_branch2c"
+    top: "res5b"
+    name: "res5b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5b"
+    name: "res5b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "bn5c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "scale5c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "bn5c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "scale5c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2c"
+    name: "res5c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "bn5c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "scale5c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b"
+    bottom: "res5c_branch2c"
+    top: "res5c"
+    name: "res5c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5c"
+    top: "res5c"
+    name: "res5c_relu"
+    type: "ReLU"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "res5c"
+  top: "rpn/output"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#----------------------new conv layer------------------
+layer {
+    bottom: "res5c"
+    top: "conv_new_1"
+    name: "conv_new_1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "conv_new_1"
+    name: "conv_new_1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_cls"
+    name: "rfcn_cls"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1029 #21*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_bbox"
+    name: "rfcn_bbox"
+    type: "Convolution"
+    convolution_param {
+        num_output: 392 #8*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+#--------------position sensitive RoI pooling--------------
+layer {
+    bottom: "rfcn_cls"
+    bottom: "rois"
+    top: "psroipooled_cls_rois"
+    name: "psroipooled_cls_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 21
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_cls_rois"
+    top: "cls_score"
+    name: "ave_cls_score_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+layer {
+    bottom: "rfcn_bbox"
+    bottom: "rois"
+    top: "psroipooled_loc_rois"
+    name: "psroipooled_loc_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 8
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_loc_rois"
+    top: "bbox_pred_pre"
+    name: "ave_bbox_pred_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+#-----------------------output------------------------
+layer {
+   name: "cls_prob"
+   type: "Softmax"
+   bottom: "cls_score"
+   top: "cls_prob_pre"
+}
+
+layer {
+    name: "cls_prob_reshape"
+    type: "Reshape"
+    bottom: "cls_prob_pre"
+    top: "cls_prob"
+    reshape_param {
+        shape {
+            dim: -1
+            dim: 21
+        }
+    }
+}
+
+layer {
+    name: "bbox_pred_reshape"
+    type: "Reshape"
+    bottom: "bbox_pred_pre"
+    top: "bbox_pred"
+    reshape_param {
+        shape {
+            dim: -1
+            dim: 8
+        }
+    }
+}
+
+
diff --git a/models/pascal_voc/ResNet-50/rfcn_end2end/train_agonistic.prototxt b/models/pascal_voc/ResNet-50/rfcn_end2end/train_agonistic.prototxt
new file mode 100644
index 0000000..13687cc
--- /dev/null
+++ b/models/pascal_voc/ResNet-50/rfcn_end2end/train_agonistic.prototxt
@@ -0,0 +1,3877 @@
+name: "ResNet-50"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+# ------------------------ conv1 -----------------------------
+layer {
+    bottom: "data"
+    top: "conv1"
+    name: "conv1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 7
+        pad: 3
+        stride: 2
+    }
+    param {
+        lr_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+    }
+    
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "bn_conv1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "scale_conv1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "conv1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv1"
+    top: "pool1"
+    name: "pool1"
+    type: "Pooling"
+    pooling_param {
+        kernel_size: 3
+        stride: 2
+        pool: MAX
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch1"
+    name: "res2a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "bn2a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "scale2a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "bn2a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "scale2a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "bn2a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "scale2a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2c"
+    name: "res2a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "bn2a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "scale2a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    bottom: "res2a_branch2c"
+    top: "res2a"
+    name: "res2a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2a"
+    name: "res2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "bn2b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "scale2b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "bn2b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "scale2b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2c"
+    name: "res2b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "bn2b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "scale2b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a"
+    bottom: "res2b_branch2c"
+    top: "res2b"
+    name: "res2b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2b"
+    name: "res2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "bn2c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "scale2c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "bn2c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "scale2c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2c"
+    name: "res2c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "bn2c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "scale2c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b"
+    bottom: "res2c_branch2c"
+    top: "res2c"
+    name: "res2c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res2c"
+    name: "res2c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch1"
+    name: "res3a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "bn3a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "scale3a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "bn3a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "scale3a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "bn3a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "scale3a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2c"
+    name: "res3a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "bn3a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "scale3a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    bottom: "res3a_branch2c"
+    top: "res3a"
+    name: "res3a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3a"
+    name: "res3a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3b_branch2a"
+    name: "res3b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "bn3b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "scale3b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "res3b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2b"
+    name: "res3b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "bn3b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "scale3b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "res3b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2c"
+    name: "res3b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2c"
+    top: "res3b_branch2c"
+    name: "bn3b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2c"
+    top: "res3b_branch2c"
+    name: "scale3b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a"
+    bottom: "res3b_branch2c"
+    top: "res3b"
+    name: "res3b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b"
+    top: "res3b"
+    name: "res3b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b"
+    top: "res3c_branch2a"
+    name: "res3c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "bn3c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "scale3c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "res3c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2b"
+    name: "res3c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "bn3c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "scale3c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "res3c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2c"
+    name: "res3c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2c"
+    top: "res3c_branch2c"
+    name: "bn3c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2c"
+    top: "res3c_branch2c"
+    name: "scale3c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b"
+    bottom: "res3c_branch2c"
+    top: "res3c"
+    name: "res3c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3c"
+    top: "res3c"
+    name: "res3c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c"
+    top: "res3d_branch2a"
+    name: "res3d_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "bn3d_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "scale3d_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "res3d_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2b"
+    name: "res3d_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "bn3d_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "scale3d_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "res3d_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2c"
+    name: "res3d_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2c"
+    top: "res3d_branch2c"
+    name: "bn3d_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2c"
+    top: "res3d_branch2c"
+    name: "scale3d_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c"
+    bottom: "res3d_branch2c"
+    top: "res3d"
+    name: "res3d"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3d"
+    top: "res3d"
+    name: "res3d_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d"
+    top: "res4a_branch1"
+    name: "res4a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "bn4a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "scale4a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "bn4a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "scale4a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "bn4a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "scale4a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2c"
+    name: "res4a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "bn4a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "scale4a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    bottom: "res4a_branch2c"
+    top: "res4a"
+    name: "res4a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4a"
+    name: "res4a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4b_branch2a"
+    name: "res4b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "bn4b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "scale4b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "res4b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2b"
+    name: "res4b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "bn4b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "scale4b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "res4b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2c"
+    name: "res4b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2c"
+    top: "res4b_branch2c"
+    name: "bn4b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2c"
+    top: "res4b_branch2c"
+    name: "scale4b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a"
+    bottom: "res4b_branch2c"
+    top: "res4b"
+    name: "res4b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b"
+    top: "res4b"
+    name: "res4b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b"
+    top: "res4c_branch2a"
+    name: "res4c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "bn4c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "scale4c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "res4c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2b"
+    name: "res4c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "bn4c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "scale4c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "res4c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2c"
+    name: "res4c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2c"
+    top: "res4c_branch2c"
+    name: "bn4c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2c"
+    top: "res4c_branch2c"
+    name: "scale4c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b"
+    bottom: "res4c_branch2c"
+    top: "res4c"
+    name: "res4c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4c"
+    top: "res4c"
+    name: "res4c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c"
+    top: "res4d_branch2a"
+    name: "res4d_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "bn4d_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "scale4d_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "res4d_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2b"
+    name: "res4d_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "bn4d_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "scale4d_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "res4d_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2c"
+    name: "res4d_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2c"
+    top: "res4d_branch2c"
+    name: "bn4d_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2c"
+    top: "res4d_branch2c"
+    name: "scale4d_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c"
+    bottom: "res4d_branch2c"
+    top: "res4d"
+    name: "res4d"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4d"
+    top: "res4d"
+    name: "res4d_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d"
+    top: "res4e_branch2a"
+    name: "res4e_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "bn4e_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "scale4e_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "res4e_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2b"
+    name: "res4e_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "bn4e_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "scale4e_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "res4e_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2c"
+    name: "res4e_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2c"
+    top: "res4e_branch2c"
+    name: "bn4e_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2c"
+    top: "res4e_branch2c"
+    name: "scale4e_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d"
+    bottom: "res4e_branch2c"
+    top: "res4e"
+    name: "res4e"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4e"
+    top: "res4e"
+    name: "res4e_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e"
+    top: "res4f_branch2a"
+    name: "res4f_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "bn4f_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "scale4f_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "res4f_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2b"
+    name: "res4f_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "bn4f_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "scale4f_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "res4f_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2c"
+    name: "res4f_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2c"
+    top: "res4f_branch2c"
+    name: "bn4f_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2c"
+    top: "res4f_branch2c"
+    name: "scale4f_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e"
+    bottom: "res4f_branch2c"
+    top: "res4f"
+    name: "res4f"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4f"
+    top: "res4f"
+    name: "res4f_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f"
+    top: "res5a_branch1"
+    name: "res5a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "bn5a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "scale5a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "bn5a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "scale5a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "bn5a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "scale5a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2c"
+    name: "res5a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "bn5a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "scale5a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    bottom: "res5a_branch2c"
+    top: "res5a"
+    name: "res5a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5a"
+    name: "res5a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "bn5b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "scale5b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "bn5b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "scale5b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2c"
+    name: "res5b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "bn5b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "scale5b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a"
+    bottom: "res5b_branch2c"
+    top: "res5b"
+    name: "res5b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5b"
+    name: "res5b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "bn5c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "scale5c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "bn5c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "scale5c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2c"
+    name: "res5c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "bn5c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "scale5c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b"
+    bottom: "res5c_branch2c"
+    top: "res5c"
+    name: "res5c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5c"
+    top: "res5c"
+    name: "res5c_relu"
+    type: "ReLU"
+}
+
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "res5c"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+#  top: 'rpn_scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#layer {
+#  name: 'debug-data'
+#  type: 'Python'
+#  bottom: 'data'
+#  bottom: 'rpn_rois'
+#  bottom: 'rpn_scores'
+#  python_param {
+#    module: 'rpn.debug_layer'
+#    layer: 'RPNDebugLayer'
+#  }
+#}
+
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 2"
+  }
+}
+
+#----------------------new conv layer------------------
+layer {
+    bottom: "res5c"
+    top: "conv_new_1"
+    name: "conv_new_1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "conv_new_1"
+    name: "conv_new_1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_cls"
+    name: "rfcn_cls"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1029 #21*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_bbox"
+    name: "rfcn_bbox"
+    type: "Convolution"
+    convolution_param {
+        num_output: 392 #8*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+#--------------position sensitive RoI pooling--------------
+layer {
+    bottom: "rfcn_cls"
+    bottom: "rois"
+    top: "psroipooled_cls_rois"
+    name: "psroipooled_cls_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 21
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_cls_rois"
+    top: "cls_score"
+    name: "ave_cls_score_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+layer {
+    bottom: "rfcn_bbox"
+    bottom: "rois"
+    top: "psroipooled_loc_rois"
+    name: "psroipooled_loc_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 8
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_loc_rois"
+    top: "bbox_pred"
+    name: "ave_bbox_pred_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+#-----------------------output------------------------
+layer {
+   name: "loss"
+   type: "SoftmaxWithLoss"
+   bottom: "cls_score"
+   bottom: "labels"
+   top: "loss_cls"
+   loss_weight: 1
+   propagate_down: true
+   propagate_down: false
+}
+
+layer {
+   name: "accuarcy"
+   type: "Accuracy"
+   bottom: "cls_score"
+   bottom: "labels"
+   top: "accuarcy"
+   #include: { phase: TEST }
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "loss_bbox"
+   type: "SmoothL1LossOHEM"
+   bottom: "bbox_pred"
+   bottom: "bbox_targets"
+   bottom: 'bbox_inside_weights'
+   top: "loss_bbox"
+   loss_weight: 1
+   loss_param {
+        normalization: PRE_FIXED
+        pre_fixed_normalizer: 128
+   }
+   propagate_down: true
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+    name: "silence"
+    type: "Silence"
+    bottom: "bbox_outside_weights"
+}
+
diff --git a/models/pascal_voc/ResNet-50/rfcn_end2end/train_agonistic_ohem.prototxt b/models/pascal_voc/ResNet-50/rfcn_end2end/train_agonistic_ohem.prototxt
new file mode 100644
index 0000000..7981ac8
--- /dev/null
+++ b/models/pascal_voc/ResNet-50/rfcn_end2end/train_agonistic_ohem.prototxt
@@ -0,0 +1,3946 @@
+name: "ResNet-50"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+# ------------------------ conv1 -----------------------------
+layer {
+    bottom: "data"
+    top: "conv1"
+    name: "conv1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 7
+        pad: 3
+        stride: 2
+    }
+    param {
+        lr_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+    }
+    
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "bn_conv1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "scale_conv1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "conv1"
+    top: "conv1"
+    name: "conv1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv1"
+    top: "pool1"
+    name: "pool1"
+    type: "Pooling"
+    pooling_param {
+        kernel_size: 3
+        stride: 2
+        pool: MAX
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch1"
+    name: "res2a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "bn2a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    top: "res2a_branch1"
+    name: "scale2a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "pool1"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "bn2a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "scale2a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2a"
+    name: "res2a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2a"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "bn2a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "scale2a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2b"
+    name: "res2a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a_branch2b"
+    top: "res2a_branch2c"
+    name: "res2a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "bn2a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch2c"
+    top: "res2a_branch2c"
+    name: "scale2a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a_branch1"
+    bottom: "res2a_branch2c"
+    top: "res2a"
+    name: "res2a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2a"
+    name: "res2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "bn2b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "scale2b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2a"
+    name: "res2b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2a"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "bn2b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "scale2b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2b"
+    name: "res2b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b_branch2b"
+    top: "res2b_branch2c"
+    name: "res2b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "bn2b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b_branch2c"
+    top: "res2b_branch2c"
+    name: "scale2b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2a"
+    bottom: "res2b_branch2c"
+    top: "res2b"
+    name: "res2b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2b"
+    name: "res2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2b"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "bn2c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "scale2c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2a"
+    name: "res2c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2a"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 64
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "bn2c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "scale2c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2b"
+    name: "res2c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c_branch2b"
+    top: "res2c_branch2c"
+    name: "res2c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "bn2c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c_branch2c"
+    top: "res2c_branch2c"
+    name: "scale2c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2b"
+    bottom: "res2c_branch2c"
+    top: "res2c"
+    name: "res2c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res2c"
+    name: "res2c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch1"
+    name: "res3a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "bn3a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    top: "res3a_branch1"
+    name: "scale3a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res2c"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "bn3a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "scale3a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2a"
+    name: "res3a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2a"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "bn3a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "scale3a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2b"
+    name: "res3a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a_branch2b"
+    top: "res3a_branch2c"
+    name: "res3a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "bn3a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch2c"
+    top: "res3a_branch2c"
+    name: "scale3a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a_branch1"
+    bottom: "res3a_branch2c"
+    top: "res3a"
+    name: "res3a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3a"
+    name: "res3a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3a"
+    top: "res3b_branch2a"
+    name: "res3b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "bn3b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "scale3b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2a"
+    name: "res3b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b_branch2a"
+    top: "res3b_branch2b"
+    name: "res3b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "bn3b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "scale3b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2b"
+    name: "res3b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b_branch2b"
+    top: "res3b_branch2c"
+    name: "res3b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2c"
+    top: "res3b_branch2c"
+    name: "bn3b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b_branch2c"
+    top: "res3b_branch2c"
+    name: "scale3b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3a"
+    bottom: "res3b_branch2c"
+    top: "res3b"
+    name: "res3b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3b"
+    top: "res3b"
+    name: "res3b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3b"
+    top: "res3c_branch2a"
+    name: "res3c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "bn3c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "scale3c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2a"
+    name: "res3c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c_branch2a"
+    top: "res3c_branch2b"
+    name: "res3c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "bn3c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "scale3c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2b"
+    name: "res3c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c_branch2b"
+    top: "res3c_branch2c"
+    name: "res3c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2c"
+    top: "res3c_branch2c"
+    name: "bn3c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c_branch2c"
+    top: "res3c_branch2c"
+    name: "scale3c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3b"
+    bottom: "res3c_branch2c"
+    top: "res3c"
+    name: "res3c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3c"
+    top: "res3c"
+    name: "res3c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3c"
+    top: "res3d_branch2a"
+    name: "res3d_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "bn3d_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "scale3d_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2a"
+    name: "res3d_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d_branch2a"
+    top: "res3d_branch2b"
+    name: "res3d_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 128
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "bn3d_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "scale3d_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2b"
+    name: "res3d_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d_branch2b"
+    top: "res3d_branch2c"
+    name: "res3d_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2c"
+    top: "res3d_branch2c"
+    name: "bn3d_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d_branch2c"
+    top: "res3d_branch2c"
+    name: "scale3d_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3c"
+    bottom: "res3d_branch2c"
+    top: "res3d"
+    name: "res3d"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res3d"
+    top: "res3d"
+    name: "res3d_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res3d"
+    top: "res4a_branch1"
+    name: "res4a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "bn4a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    top: "res4a_branch1"
+    name: "scale4a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res3d"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 2
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "bn4a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "scale4a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2a"
+    name: "res4a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2a"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "bn4a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "scale4a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2b"
+    name: "res4a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a_branch2b"
+    top: "res4a_branch2c"
+    name: "res4a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "bn4a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch2c"
+    top: "res4a_branch2c"
+    name: "scale4a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a_branch1"
+    bottom: "res4a_branch2c"
+    top: "res4a"
+    name: "res4a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4a"
+    name: "res4a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4a"
+    top: "res4b_branch2a"
+    name: "res4b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "bn4b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "scale4b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2a"
+    name: "res4b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b_branch2a"
+    top: "res4b_branch2b"
+    name: "res4b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "bn4b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "scale4b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2b"
+    name: "res4b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b_branch2b"
+    top: "res4b_branch2c"
+    name: "res4b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2c"
+    top: "res4b_branch2c"
+    name: "bn4b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b_branch2c"
+    top: "res4b_branch2c"
+    name: "scale4b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4a"
+    bottom: "res4b_branch2c"
+    top: "res4b"
+    name: "res4b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4b"
+    top: "res4b"
+    name: "res4b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4b"
+    top: "res4c_branch2a"
+    name: "res4c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "bn4c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "scale4c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2a"
+    name: "res4c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c_branch2a"
+    top: "res4c_branch2b"
+    name: "res4c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "bn4c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "scale4c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2b"
+    name: "res4c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c_branch2b"
+    top: "res4c_branch2c"
+    name: "res4c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2c"
+    top: "res4c_branch2c"
+    name: "bn4c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c_branch2c"
+    top: "res4c_branch2c"
+    name: "scale4c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4b"
+    bottom: "res4c_branch2c"
+    top: "res4c"
+    name: "res4c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4c"
+    top: "res4c"
+    name: "res4c_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4c"
+    top: "res4d_branch2a"
+    name: "res4d_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "bn4d_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "scale4d_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2a"
+    name: "res4d_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d_branch2a"
+    top: "res4d_branch2b"
+    name: "res4d_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "bn4d_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "scale4d_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2b"
+    name: "res4d_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d_branch2b"
+    top: "res4d_branch2c"
+    name: "res4d_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2c"
+    top: "res4d_branch2c"
+    name: "bn4d_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d_branch2c"
+    top: "res4d_branch2c"
+    name: "scale4d_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4c"
+    bottom: "res4d_branch2c"
+    top: "res4d"
+    name: "res4d"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4d"
+    top: "res4d"
+    name: "res4d_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4d"
+    top: "res4e_branch2a"
+    name: "res4e_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "bn4e_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "scale4e_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2a"
+    name: "res4e_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e_branch2a"
+    top: "res4e_branch2b"
+    name: "res4e_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "bn4e_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "scale4e_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2b"
+    name: "res4e_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e_branch2b"
+    top: "res4e_branch2c"
+    name: "res4e_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2c"
+    top: "res4e_branch2c"
+    name: "bn4e_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e_branch2c"
+    top: "res4e_branch2c"
+    name: "scale4e_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4d"
+    bottom: "res4e_branch2c"
+    top: "res4e"
+    name: "res4e"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4e"
+    top: "res4e"
+    name: "res4e_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4e"
+    top: "res4f_branch2a"
+    name: "res4f_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "bn4f_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "scale4f_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2a"
+    name: "res4f_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f_branch2a"
+    top: "res4f_branch2b"
+    name: "res4f_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 256
+        kernel_size: 3
+        pad: 1
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "bn4f_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "scale4f_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2b"
+    name: "res4f_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f_branch2b"
+    top: "res4f_branch2c"
+    name: "res4f_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2c"
+    top: "res4f_branch2c"
+    name: "bn4f_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f_branch2c"
+    top: "res4f_branch2c"
+    name: "scale4f_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4e"
+    bottom: "res4f_branch2c"
+    top: "res4f"
+    name: "res4f"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res4f"
+    top: "res4f"
+    name: "res4f_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res4f"
+    top: "res5a_branch1"
+    name: "res5a_branch1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "bn5a_branch1"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    top: "res5a_branch1"
+    name: "scale5a_branch1"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res4f"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "bn5a_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "scale5a_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2a"
+    name: "res5a_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2a"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "bn5a_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "scale5a_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2b"
+    name: "res5a_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a_branch2b"
+    top: "res5a_branch2c"
+    name: "res5a_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "bn5a_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch2c"
+    top: "res5a_branch2c"
+    name: "scale5a_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a_branch1"
+    bottom: "res5a_branch2c"
+    top: "res5a"
+    name: "res5a"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5a"
+    name: "res5a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "bn5b_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "scale5b_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2a"
+    name: "res5b_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2a"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "bn5b_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "scale5b_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2b"
+    name: "res5b_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b_branch2b"
+    top: "res5b_branch2c"
+    name: "res5b_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "bn5b_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b_branch2c"
+    top: "res5b_branch2c"
+    name: "scale5b_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5a"
+    bottom: "res5b_branch2c"
+    top: "res5b"
+    name: "res5b"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5b"
+    name: "res5b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5b"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "bn5c_branch2a"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "scale5c_branch2a"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2a"
+    name: "res5c_branch2a_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2a"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b"
+    type: "Convolution"
+    convolution_param {
+        num_output: 512
+        kernel_size: 3
+        dilation: 2
+        pad: 2
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "bn5c_branch2b"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "scale5c_branch2b"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2b"
+    name: "res5c_branch2b_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "res5c_branch2b"
+    top: "res5c_branch2c"
+    name: "res5c_branch2c"
+    type: "Convolution"
+    convolution_param {
+        num_output: 2048
+        kernel_size: 1
+        pad: 0
+        stride: 1
+        bias_term: false
+    }
+    param {
+        lr_mult: 1.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "bn5c_branch2c"
+    type: "BatchNorm"
+    batch_norm_param {
+        use_global_stats: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5c_branch2c"
+    top: "res5c_branch2c"
+    name: "scale5c_branch2c"
+    type: "Scale"
+    scale_param {
+        bias_term: true
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+    param {
+        lr_mult: 0.0
+        decay_mult: 0.0
+    }
+}
+
+layer {
+    bottom: "res5b"
+    bottom: "res5c_branch2c"
+    top: "res5c"
+    name: "res5c"
+    type: "Eltwise"
+}
+
+layer {
+    bottom: "res5c"
+    top: "res5c"
+    name: "res5c_relu"
+    type: "ReLU"
+}
+
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "res5c"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+#  top: 'rpn_scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#layer {
+#  name: 'debug-data'
+#  type: 'Python'
+#  bottom: 'data'
+#  bottom: 'rpn_rois'
+#  bottom: 'rpn_scores'
+#  python_param {
+#    module: 'rpn.debug_layer'
+#    layer: 'RPNDebugLayer'
+#  }
+#}
+
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 2"
+  }
+}
+
+#----------------------new conv layer------------------
+layer {
+    bottom: "res5c"
+    top: "conv_new_1"
+    name: "conv_new_1"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1024
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "conv_new_1"
+    name: "conv_new_1_relu"
+    type: "ReLU"
+}
+
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_cls"
+    name: "rfcn_cls"
+    type: "Convolution"
+    convolution_param {
+        num_output: 1029 #21*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+layer {
+    bottom: "conv_new_1"
+    top: "rfcn_bbox"
+    name: "rfcn_bbox"
+    type: "Convolution"
+    convolution_param {
+        num_output: 392 #8*(7^2) cls_num*(score_maps_size^2)
+        kernel_size: 1
+        pad: 0
+        weight_filler {
+            type: "gaussian"
+            std: 0.01
+        }
+        bias_filler {
+            type: "constant"
+            value: 0
+        }
+    }
+    param {
+        lr_mult: 1.0
+    }
+    param {
+        lr_mult: 2.0
+    }
+}
+
+#--------------position sensitive RoI pooling--------------
+layer {
+    bottom: "rfcn_cls"
+    bottom: "rois"
+    top: "psroipooled_cls_rois"
+    name: "psroipooled_cls_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 21
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_cls_rois"
+    top: "cls_score"
+    name: "ave_cls_score_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+layer {
+    bottom: "rfcn_bbox"
+    bottom: "rois"
+    top: "psroipooled_loc_rois"
+    name: "psroipooled_loc_rois"
+    type: "PSROIPooling"
+    psroi_pooling_param {
+        spatial_scale: 0.0625
+        output_dim: 8
+        group_size: 7
+    }
+}
+
+layer {
+    bottom: "psroipooled_loc_rois"
+    top: "bbox_pred"
+    name: "ave_bbox_pred_rois"
+    type: "Pooling"
+    pooling_param {
+        pool: AVE
+        kernel_size: 7
+        stride: 7
+    }
+}
+
+
+#--------------online hard example mining--------------
+layer {
+   name: "per_roi_loss_cls"
+   type: "SoftmaxWithLossOHEM"
+   bottom: "cls_score"
+   bottom: "labels"
+   top: "temp_loss_cls"
+   top: "temp_prob_cls"
+   top: "per_roi_loss_cls"
+   loss_weight: 0
+   loss_weight: 0
+   loss_weight: 0
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "per_roi_loss_bbox"
+   type: "SmoothL1LossOHEM"
+   bottom: "bbox_pred"
+   bottom: "bbox_targets"
+   bottom: "bbox_inside_weights"
+   top: "temp_loss_bbox"
+   top: "per_roi_loss_bbox"
+   loss_weight: 0
+   loss_weight: 0
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "per_roi_loss"
+   type: "Eltwise"
+   bottom: "per_roi_loss_cls"
+   bottom: "per_roi_loss_bbox"
+   top: "per_roi_loss"
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   bottom: "rois"
+   bottom: "per_roi_loss"
+   bottom: "labels"
+   bottom: "bbox_inside_weights"
+   top: "labels_ohem"
+   top: "bbox_loss_weights_ohem"
+   name: "annotator_detector"
+   type: "BoxAnnotatorOHEM"
+   box_annotator_ohem_param {
+        roi_per_img: 128
+        ignore_label: -1
+   }
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "silence"
+   type: "Silence"
+   bottom: "bbox_outside_weights"
+   bottom: "temp_loss_cls"
+   bottom: "temp_prob_cls"
+   bottom: "temp_loss_bbox"
+}
+
+#-----------------------output------------------------
+layer {
+   name: "loss"
+   type: "SoftmaxWithLoss"
+   bottom: "cls_score"
+   bottom: "labels_ohem"
+   top: "loss_cls"
+   loss_weight: 1
+   loss_param {
+        ignore_label: -1
+   }
+   propagate_down: true
+   propagate_down: false
+}
+
+layer {
+   name: "accuarcy"
+   type: "Accuracy"
+   bottom: "cls_score"
+   bottom: "labels_ohem"
+   top: "accuarcy"
+   #include: { phase: TEST }
+   accuracy_param {
+        ignore_label: -1
+   }
+   propagate_down: false
+   propagate_down: false
+}
+
+layer {
+   name: "loss_bbox"
+   type: "SmoothL1LossOHEM"
+   bottom: "bbox_pred"
+   bottom: "bbox_targets"
+   bottom: "bbox_loss_weights_ohem"
+   top: "loss_bbox"
+   loss_weight: 1
+   loss_param {
+        normalization: PRE_FIXED
+        pre_fixed_normalizer: 128
+   }
+   propagate_down: true
+   propagate_down: false
+   propagate_down: false
+}
+
diff --git a/models/pascal_voc/VGG16/fast_rcnn/solver.prototxt b/models/pascal_voc/VGG16/fast_rcnn/solver.prototxt
new file mode 100644
index 0000000..9449ab1
--- /dev/null
+++ b/models/pascal_voc/VGG16/fast_rcnn/solver.prototxt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/VGG16/fast_rcnn/train.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 30000
+display: 20
+average_loss: 100
+# iter_size: 1
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg16_fast_rcnn"
+#debug_info: true
diff --git a/models/pascal_voc/VGG16/fast_rcnn/test.prototxt b/models/pascal_voc/VGG16/fast_rcnn/test.prototxt
new file mode 100644
index 0000000..fec2369
--- /dev/null
+++ b/models/pascal_voc/VGG16/fast_rcnn/test.prototxt
@@ -0,0 +1,517 @@
+name: "VGG_ILSVRC_16_layers"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "rois"
+input_shape {
+  dim: 1 # to be changed on-the-fly to num ROIs
+  dim: 5 # [batch ind, x1, y1, x2, y2] zero-based indexing
+}
+
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+}
diff --git a/models/pascal_voc/VGG16/fast_rcnn/train.prototxt b/models/pascal_voc/VGG16/fast_rcnn/train.prototxt
new file mode 100644
index 0000000..2e7958f
--- /dev/null
+++ b/models/pascal_voc/VGG16/fast_rcnn/train.prototxt
@@ -0,0 +1,503 @@
+name: "VGG_ILSVRC_16_layers"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
diff --git a/models/pascal_voc/VGG16/faster_rcnn_alt_opt/faster_rcnn_test.pt b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/faster_rcnn_test.pt
new file mode 100644
index 0000000..c264bfd
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/faster_rcnn_test.pt
@@ -0,0 +1,409 @@
+name: "VGG_ILSVRC_16_layers"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  convolution_param {
+    num_output: 64
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  convolution_param {
+    num_output: 64
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2 stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  convolution_param {
+    num_output: 128
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  convolution_param {
+    num_output: 128
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2 stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  convolution_param {
+    num_output: 256
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  convolution_param {
+    num_output: 256
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  convolution_param {
+    num_output: 256
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2 stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2 stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5_3"
+  top: "rpn/output"
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  inner_product_param {
+    num_output: 21
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  inner_product_param {
+    num_output: 84
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+}
diff --git a/models/pascal_voc/VGG16/faster_rcnn_alt_opt/rpn_test.pt b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/rpn_test.pt
new file mode 100644
index 0000000..0e7b1da
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/rpn_test.pt
@@ -0,0 +1,341 @@
+name: "VGG_ILSVRC_16_layers"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  convolution_param {
+    num_output: 64
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  convolution_param {
+    num_output: 64
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2 stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  convolution_param {
+    num_output: 128
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  convolution_param {
+    num_output: 128
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2 stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  convolution_param {
+    num_output: 256
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  convolution_param {
+    num_output: 256
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  convolution_param {
+    num_output: 256
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2 stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2 stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5_3"
+  top: "rpn/output"
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  top: 'scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
diff --git a/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_fast_rcnn_solver30k40k.pt b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_fast_rcnn_solver30k40k.pt
new file mode 100644
index 0000000..82df70a
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_fast_rcnn_solver30k40k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 30000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg16_fast_rcnn"
diff --git a/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt
new file mode 100644
index 0000000..628dc85
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt
@@ -0,0 +1,542 @@
+name: "VGG_ILSVRC_16_layers"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
+
+#========= RPN ============
+# Dummy layers so that initial parameters are saved into the output net
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5_3"
+  top: "rpn/output"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "silence_rpn_cls_score"
+  type: "Silence"
+  bottom: "rpn_cls_score"
+}
+layer {
+  name: "silence_rpn_bbox_pred"
+  type: "Silence"
+  bottom: "rpn_bbox_pred"
+}
diff --git a/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_rpn_solver60k80k.pt b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_rpn_solver60k80k.pt
new file mode 100644
index 0000000..a6e29f0
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_rpn_solver60k80k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_rpn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 60000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg16_rpn"
diff --git a/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_rpn_train.pt b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_rpn_train.pt
new file mode 100644
index 0000000..e5f3c50
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_rpn_train.pt
@@ -0,0 +1,525 @@
+name: "VGG_ILSVRC_16_layers"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5_3"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RCNN ============
+# Dummy layers so that initial parameters are saved into the output net
+
+layer {
+  name: "dummy_roi_pool_conv5"
+  type: "DummyData"
+  top: "dummy_roi_pool_conv5"
+  dummy_data_param {
+    shape { dim: 1 dim: 25088 }
+    data_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "dummy_roi_pool_conv5"
+  top: "fc6"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "silence_fc7"
+  type: "Silence"
+  bottom: "fc7"
+}
diff --git a/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_fast_rcnn_solver30k40k.pt b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_fast_rcnn_solver30k40k.pt
new file mode 100644
index 0000000..861536c
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_fast_rcnn_solver30k40k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 30000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg16_fast_rcnn"
diff --git a/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt
new file mode 100644
index 0000000..e38bb2b
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt
@@ -0,0 +1,490 @@
+name: "VGG_ILSVRC_16_layers"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
+
+#========= RPN ============
+# Dummy layers so that initial parameters are saved into the output net
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5_3"
+  top: "rpn/output"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "silence_rpn_cls_score"
+  type: "Silence"
+  bottom: "rpn_cls_score"
+}
+layer {
+  name: "silence_rpn_bbox_pred"
+  type: "Silence"
+  bottom: "rpn_bbox_pred"
+}
diff --git a/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_rpn_solver60k80k.pt b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_rpn_solver60k80k.pt
new file mode 100644
index 0000000..7199df8
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_rpn_solver60k80k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_rpn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 60000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg16_rpn"
diff --git a/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_rpn_train.pt b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_rpn_train.pt
new file mode 100644
index 0000000..6d82857
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_rpn_train.pt
@@ -0,0 +1,465 @@
+name: "VGG_ILSVRC_16_layers"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5_3"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RCNN ============
+# Dummy layers so that initial parameters are saved into the output net
+
+layer {
+  name: "dummy_roi_pool_conv5"
+  type: "DummyData"
+  top: "dummy_roi_pool_conv5"
+  dummy_data_param {
+    shape { dim: 1 dim: 25088 }
+    data_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "dummy_roi_pool_conv5"
+  top: "fc6"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "silence_fc7"
+  type: "Silence"
+  bottom: "fc7"
+}
diff --git a/models/pascal_voc/VGG16/faster_rcnn_end2end/solver.prototxt b/models/pascal_voc/VGG16/faster_rcnn_end2end/solver.prototxt
new file mode 100644
index 0000000..7547cc8
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_end2end/solver.prototxt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/VGG16/faster_rcnn_end2end/train.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 50000
+display: 20
+average_loss: 100
+# iter_size: 1
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg16_faster_rcnn"
+iter_size: 2
diff --git a/models/pascal_voc/VGG16/faster_rcnn_end2end/test.prototxt b/models/pascal_voc/VGG16/faster_rcnn_end2end/test.prototxt
new file mode 100644
index 0000000..4a93820
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_end2end/test.prototxt
@@ -0,0 +1,608 @@
+name: "VGG_ILSVRC_16_layers"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5_3"
+  top: "rpn/output"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+}
diff --git a/models/pascal_voc/VGG16/faster_rcnn_end2end/train.prototxt b/models/pascal_voc/VGG16/faster_rcnn_end2end/train.prototxt
new file mode 100644
index 0000000..ebadb49
--- /dev/null
+++ b/models/pascal_voc/VGG16/faster_rcnn_end2end/train.prototxt
@@ -0,0 +1,673 @@
+name: "VGG_ILSVRC_16_layers"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+layer {
+  name: "conv1_1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_1"
+  type: "ReLU"
+  bottom: "conv1_1"
+  top: "conv1_1"
+}
+layer {
+  name: "conv1_2"
+  type: "Convolution"
+  bottom: "conv1_1"
+  top: "conv1_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 64
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu1_2"
+  type: "ReLU"
+  bottom: "conv1_2"
+  top: "conv1_2"
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "conv1_2"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv2_1"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2_1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_1"
+  type: "ReLU"
+  bottom: "conv2_1"
+  top: "conv2_1"
+}
+layer {
+  name: "conv2_2"
+  type: "Convolution"
+  bottom: "conv2_1"
+  top: "conv2_2"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 128
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu2_2"
+  type: "ReLU"
+  bottom: "conv2_2"
+  top: "conv2_2"
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "conv2_2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv3_1"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_1"
+  type: "ReLU"
+  bottom: "conv3_1"
+  top: "conv3_1"
+}
+layer {
+  name: "conv3_2"
+  type: "Convolution"
+  bottom: "conv3_1"
+  top: "conv3_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_2"
+  type: "ReLU"
+  bottom: "conv3_2"
+  top: "conv3_2"
+}
+layer {
+  name: "conv3_3"
+  type: "Convolution"
+  bottom: "conv3_2"
+  top: "conv3_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3_3"
+  type: "ReLU"
+  bottom: "conv3_3"
+  top: "conv3_3"
+}
+layer {
+  name: "pool3"
+  type: "Pooling"
+  bottom: "conv3_3"
+  top: "pool3"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv4_1"
+  type: "Convolution"
+  bottom: "pool3"
+  top: "conv4_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_1"
+  type: "ReLU"
+  bottom: "conv4_1"
+  top: "conv4_1"
+}
+layer {
+  name: "conv4_2"
+  type: "Convolution"
+  bottom: "conv4_1"
+  top: "conv4_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_2"
+  type: "ReLU"
+  bottom: "conv4_2"
+  top: "conv4_2"
+}
+layer {
+  name: "conv4_3"
+  type: "Convolution"
+  bottom: "conv4_2"
+  top: "conv4_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4_3"
+  type: "ReLU"
+  bottom: "conv4_3"
+  top: "conv4_3"
+}
+layer {
+  name: "pool4"
+  type: "Pooling"
+  bottom: "conv4_3"
+  top: "pool4"
+  pooling_param {
+    pool: MAX
+    kernel_size: 2
+    stride: 2
+  }
+}
+layer {
+  name: "conv5_1"
+  type: "Convolution"
+  bottom: "pool4"
+  top: "conv5_1"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_1"
+  type: "ReLU"
+  bottom: "conv5_1"
+  top: "conv5_1"
+}
+layer {
+  name: "conv5_2"
+  type: "Convolution"
+  bottom: "conv5_1"
+  top: "conv5_2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_2"
+  type: "ReLU"
+  bottom: "conv5_2"
+  top: "conv5_2"
+}
+layer {
+  name: "conv5_3"
+  type: "Convolution"
+  bottom: "conv5_2"
+  top: "conv5_3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5_3"
+  type: "ReLU"
+  bottom: "conv5_3"
+  top: "conv5_3"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5_3"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 512
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+#  top: 'rpn_scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#layer {
+#  name: 'debug-data'
+#  type: 'Python'
+#  bottom: 'data'
+#  bottom: 'rpn_rois'
+#  bottom: 'rpn_scores'
+#  python_param {
+#    module: 'rpn.debug_layer'
+#    layer: 'RPNDebugLayer'
+#  }
+#}
+
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5_3"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 7
+    pooled_h: 7
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
diff --git a/models/pascal_voc/VGG_CNN_M_1024/fast_rcnn/solver.prototxt b/models/pascal_voc/VGG_CNN_M_1024/fast_rcnn/solver.prototxt
new file mode 100644
index 0000000..eaa94d9
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/fast_rcnn/solver.prototxt
@@ -0,0 +1,15 @@
+train_net: "models/pascal_voc/VGG_CNN_M_1024/fast_rcnn/train.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 30000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg_cnn_m_1024_fast_rcnn"
+#debug_info: true
diff --git a/models/pascal_voc/VGG_CNN_M_1024/fast_rcnn/test.prototxt b/models/pascal_voc/VGG_CNN_M_1024/fast_rcnn/test.prototxt
new file mode 100644
index 0000000..baeac36
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/fast_rcnn/test.prototxt
@@ -0,0 +1,317 @@
+name: "VGG_CNN_M_1024"
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+input: "rois"
+input_shape {
+  dim: 1 # to be changed on-the-fly to num ROIs
+  dim: 5 # [batch ind, x1, y1, x2, y2] zero-based indexing
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7
+    stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 5
+    stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+}
diff --git a/models/pascal_voc/VGG_CNN_M_1024/fast_rcnn/train.prototxt b/models/pascal_voc/VGG_CNN_M_1024/fast_rcnn/train.prototxt
new file mode 100644
index 0000000..d702367
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/fast_rcnn/train.prototxt
@@ -0,0 +1,310 @@
+name: "VGG_CNN_M_1024"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7
+    stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 5
+    stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/faster_rcnn_test.pt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/faster_rcnn_test.pt
new file mode 100644
index 0000000..954b276
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/faster_rcnn_test.pt
@@ -0,0 +1,289 @@
+name: "VGG_CNN_M_1024"
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  convolution_param {
+    num_output: 96
+    kernel_size: 7
+    stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 5
+    stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  inner_product_param {
+    num_output: 21
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  inner_product_param {
+    num_output: 84
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+}
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/rpn_test.pt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/rpn_test.pt
new file mode 100644
index 0000000..fdf373f
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/rpn_test.pt
@@ -0,0 +1,221 @@
+name: "VGG_CNN_M_1024"
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  convolution_param {
+    num_output: 96
+    kernel_size: 7
+    stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 5
+    stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  top: 'scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_fast_rcnn_solver30k40k.pt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_fast_rcnn_solver30k40k.pt
new file mode 100644
index 0000000..8444a3e
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_fast_rcnn_solver30k40k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 30000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg_cnn_m_1024_fast_rcnn"
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt
new file mode 100644
index 0000000..1bc1534
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt
@@ -0,0 +1,337 @@
+name: "VGG_CNN_M_1024"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7 stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3 stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  convolution_param {
+    num_output: 256
+    pad: 1 kernel_size: 5 stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3 stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
+
+#========= RPN ============
+# Dummy layers so that initial parameters are saved into the output net
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "silence_rpn_cls_score"
+  type: "Silence"
+  bottom: "rpn_cls_score"
+}
+layer {
+  name: "silence_rpn_bbox_pred"
+  type: "Silence"
+  bottom: "rpn_bbox_pred"
+}
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_rpn_solver60k80k.pt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_rpn_solver60k80k.pt
new file mode 100644
index 0000000..6bea5fc
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_rpn_solver60k80k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_rpn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 60000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg_cnn_m_1024_rpn"
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_rpn_train.pt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_rpn_train.pt
new file mode 100644
index 0000000..29fdfe0
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage1_rpn_train.pt
@@ -0,0 +1,286 @@
+name: "VGG_CNN_M_1024"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7 stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3 stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  convolution_param {
+    num_output: 256
+    pad: 1 kernel_size: 5 stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3 stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "dummy_roi_pool_conv5"
+  type: "DummyData"
+  top: "dummy_roi_pool_conv5"
+  dummy_data_param {
+    shape { dim: 1 dim: 18432 }
+    data_filler { type: "gaussian" std: 0.01 }
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "dummy_roi_pool_conv5"
+  top: "fc6"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "silence_fc7"
+  type: "Silence"
+  bottom: "fc7"
+}
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_fast_rcnn_solver30k40k.pt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_fast_rcnn_solver30k40k.pt
new file mode 100644
index 0000000..a45a6ee
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_fast_rcnn_solver30k40k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 30000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg_cnn_m_1024_fast_rcnn"
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt
new file mode 100644
index 0000000..4825b1b
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt
@@ -0,0 +1,337 @@
+name: "VGG_CNN_M_1024"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7 stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3 stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    pad: 1 kernel_size: 5 stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3 stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param { lr_mult: 1 }
+  param { lr_mult: 2 }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
+
+#========= RPN ============
+# Dummy layers so that initial parameters are saved into the output net
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "silence_rpn_cls_score"
+  type: "Silence"
+  bottom: "rpn_cls_score"
+}
+layer {
+  name: "silence_rpn_bbox_pred"
+  type: "Silence"
+  bottom: "rpn_bbox_pred"
+}
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_rpn_solver60k80k.pt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_rpn_solver60k80k.pt
new file mode 100644
index 0000000..85f4f0c
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_rpn_solver60k80k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_rpn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 60000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg_cnn_m_1024_rpn"
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_rpn_train.pt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_rpn_train.pt
new file mode 100644
index 0000000..d27e76b
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_alt_opt/stage2_rpn_train.pt
@@ -0,0 +1,286 @@
+name: "VGG_CNN_M_1024"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7 stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3 stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    pad: 1 kernel_size: 5 stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3 stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 512
+    pad: 1 kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "dummy_roi_pool_conv5"
+  type: "DummyData"
+  top: "dummy_roi_pool_conv5"
+  dummy_data_param {
+    shape { dim: 1 dim: 18432 }
+    data_filler { type: "gaussian" std: 0.01 }
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "dummy_roi_pool_conv5"
+  top: "fc6"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "silence_fc7"
+  type: "Silence"
+  bottom: "fc7"
+}
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_end2end/solver.prototxt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_end2end/solver.prototxt
new file mode 100644
index 0000000..8134d4e
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_end2end/solver.prototxt
@@ -0,0 +1,14 @@
+train_net: "models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_end2end/train.prototxt"
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 50000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "vgg_cnn_m_1024_faster_rcnn"
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_end2end/test.prototxt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_end2end/test.prototxt
new file mode 100644
index 0000000..c8bc90a
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_end2end/test.prototxt
@@ -0,0 +1,450 @@
+name: "VGG_CNN_M_1024"
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7
+    stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 5
+    stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+#layer {
+#  name: "rpn_conv/3x3"
+#  type: "Convolution"
+#  bottom: "conv5"
+#  top: "rpn_conv/3x3"
+#  param { lr_mult: 1.0 decay_mult: 1.0 }
+#  param { lr_mult: 2.0 decay_mult: 0 }
+#  convolution_param {
+#    num_output: 192
+#    kernel_size: 3 pad: 1 stride: 1
+#    weight_filler { type: "gaussian" std: 0.01 }
+#    bias_filler { type: "constant" value: 0 }
+#  }
+#}
+#layer {
+#  name: "rpn_conv/5x5"
+#  type: "Convolution"
+#  bottom: "conv5"
+#  top: "rpn_conv/5x5"
+#  param { lr_mult: 1.0 decay_mult: 1.0 }
+#  param { lr_mult: 2.0 decay_mult: 0 }
+#  convolution_param {
+#    num_output: 64
+#    kernel_size: 5 pad: 2 stride: 1
+#    weight_filler { type: "gaussian" std: 0.0036 }
+#    bias_filler { type: "constant" value: 0 }
+#  }
+#}
+#layer {
+#  name: "rpn/output"
+#  type: "Concat"
+#  bottom: "rpn_conv/3x3"
+#  bottom: "rpn_conv/5x5"
+#  top: "rpn/output"
+#}
+#layer {
+#  name: "rpn_relu/output"
+#  type: "ReLU"
+#  bottom: "rpn/output"
+#  top: "rpn/output"
+#}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 decay_mult: 1.0 }
+  param { lr_mult: 2.0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+    decay_mult: 1
+  }
+  param {
+    lr_mult: 2
+    decay_mult: 0
+  }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+}
diff --git a/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_end2end/train.prototxt b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_end2end/train.prototxt
new file mode 100644
index 0000000..81a4d3e
--- /dev/null
+++ b/models/pascal_voc/VGG_CNN_M_1024/faster_rcnn_end2end/train.prototxt
@@ -0,0 +1,484 @@
+name: "VGG_CNN_M_1024"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+layer {
+  name: "conv1"
+  type: "Convolution"
+  bottom: "data"
+  top: "conv1"
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  param {
+    lr_mult: 0
+    decay_mult: 0
+  }
+  convolution_param {
+    num_output: 96
+    kernel_size: 7
+    stride: 2
+  }
+}
+layer {
+  name: "relu1"
+  type: "ReLU"
+  bottom: "conv1"
+  top: "conv1"
+}
+layer {
+  name: "norm1"
+  type: "LRN"
+  bottom: "conv1"
+  top: "norm1"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool1"
+  type: "Pooling"
+  bottom: "norm1"
+  top: "pool1"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv2"
+  type: "Convolution"
+  bottom: "pool1"
+  top: "conv2"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 256
+    pad: 1
+    kernel_size: 5
+    stride: 2
+  }
+}
+layer {
+  name: "relu2"
+  type: "ReLU"
+  bottom: "conv2"
+  top: "conv2"
+}
+layer {
+  name: "norm2"
+  type: "LRN"
+  bottom: "conv2"
+  top: "norm2"
+  lrn_param {
+    local_size: 5
+    alpha: 0.0005
+    beta: 0.75
+    k: 2
+  }
+}
+layer {
+  name: "pool2"
+  type: "Pooling"
+  bottom: "norm2"
+  top: "pool2"
+  pooling_param {
+    pool: MAX
+    kernel_size: 3
+    stride: 2
+  }
+}
+layer {
+  name: "conv3"
+  type: "Convolution"
+  bottom: "pool2"
+  top: "conv3"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu3"
+  type: "ReLU"
+  bottom: "conv3"
+  top: "conv3"
+}
+layer {
+  name: "conv4"
+  type: "Convolution"
+  bottom: "conv3"
+  top: "conv4"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu4"
+  type: "ReLU"
+  bottom: "conv4"
+  top: "conv4"
+}
+layer {
+  name: "conv5"
+  type: "Convolution"
+  bottom: "conv4"
+  top: "conv5"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  convolution_param {
+    num_output: 512
+    pad: 1
+    kernel_size: 3
+  }
+}
+layer {
+  name: "relu5"
+  type: "ReLU"
+  bottom: "conv5"
+  top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+#  top: 'rpn_scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#layer {
+#  name: 'debug-data'
+#  type: 'Python'
+#  bottom: 'data'
+#  bottom: 'rpn_rois'
+#  bottom: 'rpn_scores'
+#  python_param {
+#    module: 'rpn.debug_layer'
+#    layer: 'RPNDebugLayer'
+#  }
+#}
+
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "pool5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "pool5"
+  top: "fc6"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 1024
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param {
+    lr_mult: 1
+  }
+  param {
+    lr_mult: 2
+  }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "loss_cls"
+  loss_weight: 1
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "loss_bbox"
+  loss_weight: 1
+}
diff --git a/models/pascal_voc/ZF/fast_rcnn/solver.prototxt b/models/pascal_voc/ZF/fast_rcnn/solver.prototxt
new file mode 100644
index 0000000..3b346cc
--- /dev/null
+++ b/models/pascal_voc/ZF/fast_rcnn/solver.prototxt
@@ -0,0 +1,18 @@
+train_net: "models/pascal_voc/ZF/fast_rcnn/train.prototxt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 30000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "zf_fast_rcnn"
+#debug_info: true
+#iter_size: 2
diff --git a/models/pascal_voc/ZF/fast_rcnn/test.prototxt b/models/pascal_voc/ZF/fast_rcnn/test.prototxt
new file mode 100644
index 0000000..18d5cd8
--- /dev/null
+++ b/models/pascal_voc/ZF/fast_rcnn/test.prototxt
@@ -0,0 +1,251 @@
+name: "ZF"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "rois"
+input_shape {
+  dim: 1 # to be changed on-the-fly to num ROIs
+  dim: 5 # [batch ind, x1, y1, x2, y2] zero-based indexing
+}
+
+#========= conv1-conv5 ============
+
+layer {
+	name: "conv1"
+	type: "Convolution"
+	bottom: "data"
+	top: "conv1"
+	convolution_param {
+		num_output: 96
+		kernel_size: 7
+		pad: 3
+		stride: 2
+	}
+}
+layer {
+	name: "relu1"
+	type: "ReLU"
+	bottom: "conv1"
+	top: "conv1"
+}
+layer {
+	name: "norm1"
+	type: "LRN"
+	bottom: "conv1"
+	top: "norm1"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool1"
+	type: "Pooling"
+	bottom: "norm1"
+	top: "pool1"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv2"
+	type: "Convolution"
+	bottom: "pool1"
+	top: "conv2"
+	convolution_param {
+		num_output: 256
+		kernel_size: 5
+		pad: 2
+		stride: 2
+	}
+}
+layer {
+	name: "relu2"
+	type: "ReLU"
+	bottom: "conv2"
+	top: "conv2"
+}
+layer {
+	name: "norm2"
+	type: "LRN"
+	bottom: "conv2"
+	top: "norm2"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool2"
+	type: "Pooling"
+	bottom: "norm2"
+	top: "pool2"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv3"
+	type: "Convolution"
+	bottom: "pool2"
+	top: "conv3"
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu3"
+	type: "ReLU"
+	bottom: "conv3"
+	top: "conv3"
+}
+layer {
+	name: "conv4"
+	type: "Convolution"
+	bottom: "conv3"
+	top: "conv4"
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu4"
+	type: "ReLU"
+	bottom: "conv4"
+	top: "conv4"
+}
+layer {
+	name: "conv5"
+	type: "Convolution"
+	bottom: "conv4"
+	top: "conv5"
+	convolution_param {
+		num_output: 256
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu5"
+	type: "ReLU"
+	bottom: "conv5"
+	top: "conv5"
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool_conv5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "roi_pool_conv5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "roi_pool_conv5"
+  top: "fc6"
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  inner_product_param {
+    num_output: 21
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  inner_product_param {
+    num_output: 84
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
diff --git a/models/pascal_voc/ZF/fast_rcnn/train.prototxt b/models/pascal_voc/ZF/fast_rcnn/train.prototxt
new file mode 100644
index 0000000..921d8e3
--- /dev/null
+++ b/models/pascal_voc/ZF/fast_rcnn/train.prototxt
@@ -0,0 +1,300 @@
+name: "ZF"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#========= conv1-conv5 ============
+
+layer {
+	name: "conv1"
+	type: "Convolution"
+	bottom: "data"
+	top: "conv1"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 96
+		kernel_size: 7
+		pad: 3
+		stride: 2
+	}
+}
+layer {
+	name: "relu1"
+	type: "ReLU"
+	bottom: "conv1"
+	top: "conv1"
+}
+layer {
+	name: "norm1"
+	type: "LRN"
+	bottom: "conv1"
+	top: "norm1"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool1"
+	type: "Pooling"
+	bottom: "norm1"
+	top: "pool1"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv2"
+	type: "Convolution"
+	bottom: "pool1"
+	top: "conv2"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 5
+		pad: 2
+		stride: 2
+	}
+}
+layer {
+	name: "relu2"
+	type: "ReLU"
+	bottom: "conv2"
+	top: "conv2"
+}
+layer {
+	name: "norm2"
+	type: "LRN"
+	bottom: "conv2"
+	top: "norm2"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool2"
+	type: "Pooling"
+	bottom: "norm2"
+	top: "pool2"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv3"
+	type: "Convolution"
+	bottom: "pool2"
+	top: "conv3"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu3"
+	type: "ReLU"
+	bottom: "conv3"
+	top: "conv3"
+}
+layer {
+	name: "conv4"
+	type: "Convolution"
+	bottom: "conv3"
+	top: "conv4"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu4"
+	type: "ReLU"
+	bottom: "conv4"
+	top: "conv4"
+}
+layer {
+	name: "conv5"
+	type: "Convolution"
+	bottom: "conv4"
+	top: "conv5"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu5"
+	type: "ReLU"
+	bottom: "conv5"
+	top: "conv5"
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool_conv5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "roi_pool_conv5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "roi_pool_conv5"
+  top: "fc6"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "bbox_loss"
+  loss_weight: 1
+}
diff --git a/models/pascal_voc/ZF/faster_rcnn_alt_opt/faster_rcnn_test.pt b/models/pascal_voc/ZF/faster_rcnn_alt_opt/faster_rcnn_test.pt
new file mode 100644
index 0000000..b24aae4
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_alt_opt/faster_rcnn_test.pt
@@ -0,0 +1,327 @@
+name: "ZF"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+#========= conv1-conv5 ============
+
+layer {
+	name: "conv1"
+	type: "Convolution"
+	bottom: "data"
+	top: "conv1"
+	convolution_param {
+		num_output: 96
+		kernel_size: 7
+		pad: 3
+		stride: 2
+	}
+}
+layer {
+	name: "relu1"
+	type: "ReLU"
+	bottom: "conv1"
+	top: "conv1"
+}
+layer {
+	name: "norm1"
+	type: "LRN"
+	bottom: "conv1"
+	top: "norm1"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool1"
+	type: "Pooling"
+	bottom: "norm1"
+	top: "pool1"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv2"
+	type: "Convolution"
+	bottom: "pool1"
+	top: "conv2"
+	convolution_param {
+		num_output: 256
+		kernel_size: 5
+		pad: 2
+		stride: 2
+	}
+}
+layer {
+	name: "relu2"
+	type: "ReLU"
+	bottom: "conv2"
+	top: "conv2"
+}
+layer {
+	name: "norm2"
+	type: "LRN"
+	bottom: "conv2"
+	top: "norm2"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool2"
+	type: "Pooling"
+	bottom: "norm2"
+	top: "pool2"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv3"
+	type: "Convolution"
+	bottom: "pool2"
+	top: "conv3"
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu3"
+	type: "ReLU"
+	bottom: "conv3"
+	top: "conv3"
+}
+layer {
+	name: "conv4"
+	type: "Convolution"
+	bottom: "conv3"
+	top: "conv4"
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu4"
+	type: "ReLU"
+	bottom: "conv4"
+	top: "conv4"
+}
+layer {
+	name: "conv5"
+	type: "Convolution"
+	bottom: "conv4"
+	top: "conv5"
+	convolution_param {
+		num_output: 256
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu5"
+	type: "ReLU"
+	bottom: "conv5"
+	top: "conv5"
+}
+
+#========= RPN ============
+
+
+layer {
+  name: "rpn_conv1"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn_conv1"
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+  }
+}
+layer {
+  name: "rpn_relu1"
+  type: "ReLU"
+  bottom: "rpn_conv1"
+  top: "rpn_conv1"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_cls_score"
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_bbox_pred"
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool_conv5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "roi_pool_conv5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "roi_pool_conv5"
+  top: "fc6"
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  inner_product_param {
+    num_output: 21
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  inner_product_param {
+    num_output: 84
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
diff --git a/models/pascal_voc/ZF/faster_rcnn_alt_opt/rpn_test.pt b/models/pascal_voc/ZF/faster_rcnn_alt_opt/rpn_test.pt
new file mode 100644
index 0000000..204f08f
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_alt_opt/rpn_test.pt
@@ -0,0 +1,233 @@
+name: "ZF"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+# ------------------------ layer 1 -----------------------------
+layer {
+	name: "conv1"
+	type: "Convolution"
+	bottom: "data"
+	top: "conv1"
+	convolution_param {
+		num_output: 96
+		kernel_size: 7
+		pad: 3
+		stride: 2
+	}
+}
+layer {
+	name: "relu1"
+	type: "ReLU"
+	bottom: "conv1"
+	top: "conv1"
+}
+layer {
+	name: "norm1"
+	type: "LRN"
+	bottom: "conv1"
+	top: "norm1"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool1"
+	type: "Pooling"
+	bottom: "norm1"
+	top: "pool1"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv2"
+	type: "Convolution"
+	bottom: "pool1"
+	top: "conv2"
+	convolution_param {
+		num_output: 256
+		kernel_size: 5
+		pad: 2
+		stride: 2
+	}
+}
+layer {
+	name: "relu2"
+	type: "ReLU"
+	bottom: "conv2"
+	top: "conv2"
+}
+
+layer {
+	name: "norm2"
+	type: "LRN"
+	bottom: "conv2"
+	top: "norm2"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool2"
+	type: "Pooling"
+	bottom: "norm2"
+	top: "pool2"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv3"
+	type: "Convolution"
+	bottom: "pool2"
+	top: "conv3"
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu3"
+	type: "ReLU"
+	bottom: "conv3"
+	top: "conv3"
+}
+layer {
+	name: "conv4"
+	type: "Convolution"
+	bottom: "conv3"
+	top: "conv4"
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu4"
+	type: "ReLU"
+	bottom: "conv4"
+	top: "conv4"
+}
+layer {
+	name: "conv5"
+	type: "Convolution"
+	bottom: "conv4"
+	top: "conv5"
+	convolution_param {
+		num_output: 256
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu5"
+	type: "ReLU"
+	bottom: "conv5"
+	top: "conv5"
+}
+
+#-----------------------layer +-------------------------
+
+layer {
+  name: "rpn_conv1"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn_conv1"
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+  }
+}
+layer {
+  name: "rpn_relu1"
+  type: "ReLU"
+  bottom: "rpn_conv1"
+  top: "rpn_conv1"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_cls_score"
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_bbox_pred"
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#-----------------------output------------------------
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  top: 'scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
diff --git a/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_fast_rcnn_solver30k40k.pt b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_fast_rcnn_solver30k40k.pt
new file mode 100644
index 0000000..0180e7c
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_fast_rcnn_solver30k40k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 30000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "zf_fast_rcnn"
diff --git a/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt
new file mode 100644
index 0000000..3d98184
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt
@@ -0,0 +1,362 @@
+name: "ZF"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#========= conv1-conv5 ============
+
+layer {
+	name: "conv1"
+	type: "Convolution"
+	bottom: "data"
+	top: "conv1"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 96
+		kernel_size: 7
+		pad: 3
+		stride: 2
+	}
+}
+layer {
+	name: "relu1"
+	type: "ReLU"
+	bottom: "conv1"
+	top: "conv1"
+}
+layer {
+	name: "norm1"
+	type: "LRN"
+	bottom: "conv1"
+	top: "norm1"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool1"
+	type: "Pooling"
+	bottom: "norm1"
+	top: "pool1"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv2"
+	type: "Convolution"
+	bottom: "pool1"
+	top: "conv2"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 5
+		pad: 2
+		stride: 2
+	}
+}
+layer {
+	name: "relu2"
+	type: "ReLU"
+	bottom: "conv2"
+	top: "conv2"
+}
+layer {
+	name: "norm2"
+	type: "LRN"
+	bottom: "conv2"
+	top: "norm2"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool2"
+	type: "Pooling"
+	bottom: "norm2"
+	top: "pool2"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv3"
+	type: "Convolution"
+	bottom: "pool2"
+	top: "conv3"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu3"
+	type: "ReLU"
+	bottom: "conv3"
+	top: "conv3"
+}
+layer {
+	name: "conv4"
+	type: "Convolution"
+	bottom: "conv3"
+	top: "conv4"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu4"
+	type: "ReLU"
+	bottom: "conv4"
+	top: "conv4"
+}
+layer {
+	name: "conv5"
+	type: "Convolution"
+	bottom: "conv4"
+	top: "conv5"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu5"
+	type: "ReLU"
+	bottom: "conv5"
+	top: "conv5"
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool_conv5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "roi_pool_conv5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "roi_pool_conv5"
+  top: "fc6"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "bbox_loss"
+  loss_weight: 1
+}
+
+#========= RPN ============
+# Dummy layers so that initial parameters are saved into the output net
+
+layer {
+  name: "rpn_conv1"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn_conv1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu1"
+  type: "ReLU"
+  bottom: "rpn_conv1"
+  top: "rpn_conv1"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_cls_score"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "silence_rpn_cls_score"
+  type: "Silence"
+  bottom: "rpn_cls_score"
+}
+layer {
+  name: "silence_rpn_bbox_pred"
+  type: "Silence"
+  bottom: "rpn_bbox_pred"
+}
diff --git a/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_rpn_solver60k80k.pt b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_rpn_solver60k80k.pt
new file mode 100644
index 0000000..23a7c6a
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_rpn_solver60k80k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_rpn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 60000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "zf_rpn"
diff --git a/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_rpn_train.pt b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_rpn_train.pt
new file mode 100644
index 0000000..adf8605
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage1_rpn_train.pt
@@ -0,0 +1,312 @@
+name: "ZF"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#========= conv1-conv5 ============
+
+layer {
+	name: "conv1"
+	type: "Convolution"
+	bottom: "data"
+	top: "conv1"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 96
+		kernel_size: 7
+		pad: 3
+		stride: 2
+	}
+}
+layer {
+	name: "relu1"
+	type: "ReLU"
+	bottom: "conv1"
+	top: "conv1"
+}
+layer {
+	name: "norm1"
+	type: "LRN"
+	bottom: "conv1"
+	top: "norm1"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool1"
+	type: "Pooling"
+	bottom: "norm1"
+	top: "pool1"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv2"
+	type: "Convolution"
+	bottom: "pool1"
+	top: "conv2"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 5
+		pad: 2
+		stride: 2
+	}
+}
+layer {
+	name: "relu2"
+	type: "ReLU"
+	bottom: "conv2"
+	top: "conv2"
+}
+layer {
+	name: "norm2"
+	type: "LRN"
+	bottom: "conv2"
+	top: "norm2"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool2"
+	type: "Pooling"
+	bottom: "norm2"
+	top: "pool2"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv3"
+	type: "Convolution"
+	bottom: "pool2"
+	top: "conv3"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu3"
+	type: "ReLU"
+	bottom: "conv3"
+	top: "conv3"
+}
+layer {
+	name: "conv4"
+	type: "Convolution"
+	bottom: "conv3"
+	top: "conv4"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu4"
+	type: "ReLU"
+	bottom: "conv4"
+	top: "conv4"
+}
+layer {
+	name: "conv5"
+	type: "Convolution"
+	bottom: "conv4"
+	top: "conv5"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu5"
+	type: "ReLU"
+	bottom: "conv5"
+	top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv1"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn_conv1"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu1"
+  type: "ReLU"
+  bottom: "rpn_conv1"
+  top: "rpn_conv1"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: "rpn_bbox_inside_weights"
+  bottom: "rpn_bbox_outside_weights"
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RCNN ============
+# Dummy layers so that initial parameters are saved into the output net
+
+layer {
+  name: "dummy_roi_pool_conv5"
+  type: "DummyData"
+  top: "dummy_roi_pool_conv5"
+  dummy_data_param {
+    shape { dim: 1 dim: 9216 }
+    data_filler { type: "gaussian" std: 0.01 }
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "dummy_roi_pool_conv5"
+  top: "fc6"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "silence_fc7"
+  type: "Silence"
+  bottom: "fc7"
+}
diff --git a/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_fast_rcnn_solver30k40k.pt b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_fast_rcnn_solver30k40k.pt
new file mode 100644
index 0000000..a666def
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_fast_rcnn_solver30k40k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 30000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "zf_fast_rcnn"
diff --git a/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt
new file mode 100644
index 0000000..262ed65
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt
@@ -0,0 +1,362 @@
+name: "ZF"
+layer {
+  name: 'data'
+  type: 'Python'
+  top: 'data'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#========= conv1-conv5 ============
+
+layer {
+	name: "conv1"
+	type: "Convolution"
+	bottom: "data"
+	top: "conv1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+	convolution_param {
+		num_output: 96
+		kernel_size: 7
+		pad: 3
+		stride: 2
+	}
+}
+layer {
+	name: "relu1"
+	type: "ReLU"
+	bottom: "conv1"
+	top: "conv1"
+}
+layer {
+	name: "norm1"
+	type: "LRN"
+	bottom: "conv1"
+	top: "norm1"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool1"
+	type: "Pooling"
+	bottom: "norm1"
+	top: "pool1"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv2"
+	type: "Convolution"
+	bottom: "pool1"
+	top: "conv2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 5
+		pad: 2
+		stride: 2
+	}
+}
+layer {
+	name: "relu2"
+	type: "ReLU"
+	bottom: "conv2"
+	top: "conv2"
+}
+layer {
+	name: "norm2"
+	type: "LRN"
+	bottom: "conv2"
+	top: "norm2"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool2"
+	type: "Pooling"
+	bottom: "norm2"
+	top: "pool2"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv3"
+	type: "Convolution"
+	bottom: "pool2"
+	top: "conv3"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu3"
+	type: "ReLU"
+	bottom: "conv3"
+	top: "conv3"
+}
+layer {
+	name: "conv4"
+	type: "Convolution"
+	bottom: "conv3"
+	top: "conv4"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu4"
+	type: "ReLU"
+	bottom: "conv4"
+	top: "conv4"
+}
+layer {
+	name: "conv5"
+	type: "Convolution"
+	bottom: "conv4"
+	top: "conv5"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu5"
+	type: "ReLU"
+	bottom: "conv5"
+	top: "conv5"
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool_conv5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "roi_pool_conv5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "roi_pool_conv5"
+  top: "fc6"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: "bbox_inside_weights"
+  bottom: "bbox_outside_weights"
+  top: "bbox_loss"
+  loss_weight: 1
+}
+
+#========= RPN ============
+# Dummy layers so that initial parameters are saved into the output net
+
+layer {
+  name: "rpn_conv1"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn_conv1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu1"
+  type: "ReLU"
+  bottom: "rpn_conv1"
+  top: "rpn_conv1"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_cls_score"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "silence_rpn_cls_score"
+  type: "Silence"
+  bottom: "rpn_cls_score"
+}
+layer {
+  name: "silence_rpn_bbox_pred"
+  type: "Silence"
+  bottom: "rpn_bbox_pred"
+}
diff --git a/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_rpn_solver60k80k.pt b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_rpn_solver60k80k.pt
new file mode 100644
index 0000000..15d3da7
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_rpn_solver60k80k.pt
@@ -0,0 +1,16 @@
+train_net: "models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_rpn_train.pt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 60000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "zf_rpn"
diff --git a/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_rpn_train.pt b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_rpn_train.pt
new file mode 100644
index 0000000..336b05b
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_alt_opt/stage2_rpn_train.pt
@@ -0,0 +1,312 @@
+name: "ZF"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#========= conv1-conv5 ============
+
+layer {
+	name: "conv1"
+	type: "Convolution"
+	bottom: "data"
+	top: "conv1"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+	convolution_param {
+		num_output: 96
+		kernel_size: 7
+		pad: 3
+		stride: 2
+	}
+}
+layer {
+	name: "relu1"
+	type: "ReLU"
+	bottom: "conv1"
+	top: "conv1"
+}
+layer {
+	name: "norm1"
+	type: "LRN"
+	bottom: "conv1"
+	top: "norm1"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool1"
+	type: "Pooling"
+	bottom: "norm1"
+	top: "pool1"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv2"
+	type: "Convolution"
+	bottom: "pool1"
+	top: "conv2"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 5
+		pad: 2
+		stride: 2
+	}
+}
+layer {
+	name: "relu2"
+	type: "ReLU"
+	bottom: "conv2"
+	top: "conv2"
+}
+layer {
+	name: "norm2"
+	type: "LRN"
+	bottom: "conv2"
+	top: "norm2"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool2"
+	type: "Pooling"
+	bottom: "norm2"
+	top: "pool2"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv3"
+	type: "Convolution"
+	bottom: "pool2"
+	top: "conv3"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu3"
+	type: "ReLU"
+	bottom: "conv3"
+	top: "conv3"
+}
+layer {
+	name: "conv4"
+	type: "Convolution"
+	bottom: "conv3"
+	top: "conv4"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu4"
+	type: "ReLU"
+	bottom: "conv4"
+	top: "conv4"
+}
+layer {
+	name: "conv5"
+	type: "Convolution"
+	bottom: "conv4"
+	top: "conv5"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu5"
+	type: "ReLU"
+	bottom: "conv5"
+	top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv1"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn_conv1"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu1"
+  type: "ReLU"
+  bottom: "rpn_conv1"
+  top: "rpn_conv1"
+}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn_conv1"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: "rpn_bbox_inside_weights"
+  bottom: "rpn_bbox_outside_weights"
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RCNN ============
+# Dummy layers so that initial parameters are saved into the output net
+
+layer {
+  name: "dummy_roi_pool_conv5"
+  type: "DummyData"
+  top: "dummy_roi_pool_conv5"
+  dummy_data_param {
+    shape { dim: 1 dim: 9216 }
+    data_filler { type: "gaussian" std: 0.01 }
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "dummy_roi_pool_conv5"
+  top: "fc6"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 0 decay_mult: 0 }
+  param { lr_mult: 0 decay_mult: 0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "silence_fc7"
+  type: "Silence"
+  bottom: "fc7"
+}
diff --git a/models/pascal_voc/ZF/faster_rcnn_end2end/solver.prototxt b/models/pascal_voc/ZF/faster_rcnn_end2end/solver.prototxt
new file mode 100644
index 0000000..246697a
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_end2end/solver.prototxt
@@ -0,0 +1,25 @@
+train_net: "models/pascal_voc/ZF/faster_rcnn_end2end/train.prototxt"
+
+base_lr: 0.001
+lr_policy: "step"
+gamma: 0.1
+stepsize: 50000
+display: 20
+average_loss: 100
+momentum: 0.9
+weight_decay: 0.0005
+
+#base_lr: 0.001
+#lr_policy: "exp"
+#gamma: 0.999539589  # (0.00001/0.001)^(1/10000)
+#display: 1
+#average_loss: 100
+#momentum: 0.9
+#weight_decay: 0.0005
+
+# We disable standard caffe solver snapshotting and implement our own snapshot
+# function
+snapshot: 0
+# We still use the snapshot prefix, though
+snapshot_prefix: "zf_faster_rcnn"
+iter_size: 2
diff --git a/models/pascal_voc/ZF/faster_rcnn_end2end/test.prototxt b/models/pascal_voc/ZF/faster_rcnn_end2end/test.prototxt
new file mode 100644
index 0000000..6d88dc3
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_end2end/test.prototxt
@@ -0,0 +1,373 @@
+name: "ZF"
+
+input: "data"
+input_shape {
+  dim: 1
+  dim: 3
+  dim: 224
+  dim: 224
+}
+
+input: "im_info"
+input_shape {
+  dim: 1
+  dim: 3
+}
+
+#========= conv1-conv5 ============
+
+layer {
+	name: "conv1"
+	type: "Convolution"
+	bottom: "data"
+	top: "conv1"
+	convolution_param {
+		num_output: 96
+		kernel_size: 7
+		pad: 3
+		stride: 2
+	}
+}
+layer {
+	name: "relu1"
+	type: "ReLU"
+	bottom: "conv1"
+	top: "conv1"
+}
+layer {
+	name: "norm1"
+	type: "LRN"
+	bottom: "conv1"
+	top: "norm1"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool1"
+	type: "Pooling"
+	bottom: "norm1"
+	top: "pool1"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv2"
+	type: "Convolution"
+	bottom: "pool1"
+	top: "conv2"
+	convolution_param {
+		num_output: 256
+		kernel_size: 5
+		pad: 2
+		stride: 2
+	}
+}
+layer {
+	name: "relu2"
+	type: "ReLU"
+	bottom: "conv2"
+	top: "conv2"
+}
+layer {
+	name: "norm2"
+	type: "LRN"
+	bottom: "conv2"
+	top: "norm2"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool2"
+	type: "Pooling"
+	bottom: "norm2"
+	top: "pool2"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv3"
+	type: "Convolution"
+	bottom: "pool2"
+	top: "conv3"
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu3"
+	type: "ReLU"
+	bottom: "conv3"
+	top: "conv3"
+}
+layer {
+	name: "conv4"
+	type: "Convolution"
+	bottom: "conv3"
+	top: "conv4"
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu4"
+	type: "ReLU"
+	bottom: "conv4"
+	top: "conv4"
+}
+layer {
+	name: "conv5"
+	type: "Convolution"
+	bottom: "conv4"
+	top: "conv5"
+	convolution_param {
+		num_output: 256
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu5"
+	type: "ReLU"
+	bottom: "conv5"
+	top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+#layer {
+#  name: "rpn_conv/3x3"
+#  type: "Convolution"
+#  bottom: "conv5"
+#  top: "rpn_conv/3x3"
+#  param { lr_mult: 1.0 decay_mult: 1.0 }
+#  param { lr_mult: 2.0 decay_mult: 0 }
+#  convolution_param {
+#    num_output: 192
+#    kernel_size: 3 pad: 1 stride: 1
+#    weight_filler { type: "gaussian" std: 0.01 }
+#    bias_filler { type: "constant" value: 0 }
+#  }
+#}
+#layer {
+#  name: "rpn_conv/5x5"
+#  type: "Convolution"
+#  bottom: "conv5"
+#  top: "rpn_conv/5x5"
+#  param { lr_mult: 1.0 decay_mult: 1.0 }
+#  param { lr_mult: 2.0 decay_mult: 0 }
+#  convolution_param {
+#    num_output: 64
+#    kernel_size: 5 pad: 2 stride: 1
+#    weight_filler { type: "gaussian" std: 0.0036 }
+#    bias_filler { type: "constant" value: 0 }
+#  }
+#}
+#layer {
+#  name: "rpn/output"
+#  type: "Concat"
+#  bottom: "rpn_conv/3x3"
+#  bottom: "rpn_conv/5x5"
+#  top: "rpn/output"
+#}
+#layer {
+#  name: "rpn_relu/output"
+#  type: "ReLU"
+#  bottom: "rpn/output"
+#  top: "rpn/output"
+#}
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rois'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool_conv5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "roi_pool_conv5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "roi_pool_conv5"
+  top: "fc6"
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  inner_product_param {
+    num_output: 21
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  inner_product_param {
+    num_output: 84
+  }
+}
+layer {
+  name: "cls_prob"
+  type: "Softmax"
+  bottom: "cls_score"
+  top: "cls_prob"
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
diff --git a/models/pascal_voc/ZF/faster_rcnn_end2end/train.prototxt b/models/pascal_voc/ZF/faster_rcnn_end2end/train.prototxt
new file mode 100644
index 0000000..c044fd5
--- /dev/null
+++ b/models/pascal_voc/ZF/faster_rcnn_end2end/train.prototxt
@@ -0,0 +1,497 @@
+name: "ZF"
+layer {
+  name: 'input-data'
+  type: 'Python'
+  top: 'data'
+  top: 'im_info'
+  top: 'gt_boxes'
+  python_param {
+    module: 'roi_data_layer.layer'
+    layer: 'RoIDataLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#========= conv1-conv5 ============
+
+layer {
+	name: "conv1"
+	type: "Convolution"
+	bottom: "data"
+	top: "conv1"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 96
+		kernel_size: 7
+		pad: 3
+		stride: 2
+	}
+}
+layer {
+	name: "relu1"
+	type: "ReLU"
+	bottom: "conv1"
+	top: "conv1"
+}
+layer {
+	name: "norm1"
+	type: "LRN"
+	bottom: "conv1"
+	top: "norm1"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool1"
+	type: "Pooling"
+	bottom: "norm1"
+	top: "pool1"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv2"
+	type: "Convolution"
+	bottom: "pool1"
+	top: "conv2"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 5
+		pad: 2
+		stride: 2
+	}
+}
+layer {
+	name: "relu2"
+	type: "ReLU"
+	bottom: "conv2"
+	top: "conv2"
+}
+layer {
+	name: "norm2"
+	type: "LRN"
+	bottom: "conv2"
+	top: "norm2"
+	lrn_param {
+		local_size: 3
+		alpha: 0.00005
+		beta: 0.75
+		norm_region: WITHIN_CHANNEL
+    engine: CAFFE
+	}
+}
+layer {
+	name: "pool2"
+	type: "Pooling"
+	bottom: "norm2"
+	top: "pool2"
+	pooling_param {
+		kernel_size: 3
+		stride: 2
+		pad: 1
+		pool: MAX
+	}
+}
+layer {
+	name: "conv3"
+	type: "Convolution"
+	bottom: "pool2"
+	top: "conv3"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu3"
+	type: "ReLU"
+	bottom: "conv3"
+	top: "conv3"
+}
+layer {
+	name: "conv4"
+	type: "Convolution"
+	bottom: "conv3"
+	top: "conv4"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 384
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu4"
+	type: "ReLU"
+	bottom: "conv4"
+	top: "conv4"
+}
+layer {
+	name: "conv5"
+	type: "Convolution"
+	bottom: "conv4"
+	top: "conv5"
+	param { lr_mult: 1.0 }
+	param { lr_mult: 2.0 }
+	convolution_param {
+		num_output: 256
+		kernel_size: 3
+		pad: 1
+		stride: 1
+	}
+}
+layer {
+	name: "relu5"
+	type: "ReLU"
+	bottom: "conv5"
+	top: "conv5"
+}
+
+#========= RPN ============
+
+layer {
+  name: "rpn_conv/3x3"
+  type: "Convolution"
+  bottom: "conv5"
+  top: "rpn/output"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 256
+    kernel_size: 3 pad: 1 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_relu/3x3"
+  type: "ReLU"
+  bottom: "rpn/output"
+  top: "rpn/output"
+}
+
+#layer {
+#  name: "rpn_conv/3x3"
+#  type: "Convolution"
+#  bottom: "conv5"
+#  top: "rpn_conv/3x3"
+#  param { lr_mult: 1.0 }
+#  param { lr_mult: 2.0 }
+#  convolution_param {
+#    num_output: 192
+#    kernel_size: 3 pad: 1 stride: 1
+#    weight_filler { type: "gaussian" std: 0.01 }
+#    bias_filler { type: "constant" value: 0 }
+#  }
+#}
+#layer {
+#  name: "rpn_conv/5x5"
+#  type: "Convolution"
+#  bottom: "conv5"
+#  top: "rpn_conv/5x5"
+#  param { lr_mult: 1.0 }
+#  param { lr_mult: 2.0 }
+#  convolution_param {
+#    num_output: 64
+#    kernel_size: 5 pad: 2 stride: 1
+#    weight_filler { type: "gaussian" std: 0.0036 }
+#    bias_filler { type: "constant" value: 0 }
+#  }
+#}
+#layer {
+#  name: "rpn/output"
+#  type: "Concat"
+#  bottom: "rpn_conv/3x3"
+#  bottom: "rpn_conv/5x5"
+#  top: "rpn/output"
+#}
+#layer {
+#  name: "rpn_relu/output"
+#  type: "ReLU"
+#  bottom: "rpn/output"
+#  top: "rpn/output"
+#}
+
+layer {
+  name: "rpn_cls_score"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 18   # 2(bg/fg) * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+  name: "rpn_bbox_pred"
+  type: "Convolution"
+  bottom: "rpn/output"
+  top: "rpn_bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  convolution_param {
+    num_output: 36   # 4 * 9(anchors)
+    kernel_size: 1 pad: 0 stride: 1
+    weight_filler { type: "gaussian" std: 0.01 }
+    bias_filler { type: "constant" value: 0 }
+  }
+}
+layer {
+   bottom: "rpn_cls_score"
+   top: "rpn_cls_score_reshape"
+   name: "rpn_cls_score_reshape"
+   type: "Reshape"
+   reshape_param { shape { dim: 0 dim: 2 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'rpn-data'
+  type: 'Python'
+  bottom: 'rpn_cls_score'
+  bottom: 'gt_boxes'
+  bottom: 'im_info'
+  bottom: 'data'
+  top: 'rpn_labels'
+  top: 'rpn_bbox_targets'
+  top: 'rpn_bbox_inside_weights'
+  top: 'rpn_bbox_outside_weights'
+  python_param {
+    module: 'rpn.anchor_target_layer'
+    layer: 'AnchorTargetLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+layer {
+  name: "rpn_loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "rpn_cls_score_reshape"
+  bottom: "rpn_labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "rpn_cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+layer {
+  name: "rpn_loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "rpn_bbox_pred"
+  bottom: "rpn_bbox_targets"
+  bottom: 'rpn_bbox_inside_weights'
+  bottom: 'rpn_bbox_outside_weights'
+  top: "rpn_loss_bbox"
+  loss_weight: 1
+  smooth_l1_loss_param { sigma: 3.0 }
+}
+
+#========= RoI Proposal ============
+
+layer {
+  name: "rpn_cls_prob"
+  type: "Softmax"
+  bottom: "rpn_cls_score_reshape"
+  top: "rpn_cls_prob"
+}
+layer {
+  name: 'rpn_cls_prob_reshape'
+  type: 'Reshape'
+  bottom: 'rpn_cls_prob'
+  top: 'rpn_cls_prob_reshape'
+  reshape_param { shape { dim: 0 dim: 18 dim: -1 dim: 0 } }
+}
+layer {
+  name: 'proposal'
+  type: 'Python'
+  bottom: 'rpn_cls_prob_reshape'
+  bottom: 'rpn_bbox_pred'
+  bottom: 'im_info'
+  top: 'rpn_rois'
+#  top: 'rpn_scores'
+  python_param {
+    module: 'rpn.proposal_layer'
+    layer: 'ProposalLayer'
+    param_str: "'feat_stride': 16"
+  }
+}
+#layer {
+#  name: 'debug-data'
+#  type: 'Python'
+#  bottom: 'data'
+#  bottom: 'rpn_rois'
+#  bottom: 'rpn_scores'
+#  python_param {
+#    module: 'rpn.debug_layer'
+#    layer: 'RPNDebugLayer'
+#  }
+#}
+layer {
+  name: 'roi-data'
+  type: 'Python'
+  bottom: 'rpn_rois'
+  bottom: 'gt_boxes'
+  top: 'rois'
+  top: 'labels'
+  top: 'bbox_targets'
+  top: 'bbox_inside_weights'
+  top: 'bbox_outside_weights'
+  python_param {
+    module: 'rpn.proposal_target_layer'
+    layer: 'ProposalTargetLayer'
+    param_str: "'num_classes': 21"
+  }
+}
+
+#========= RCNN ============
+
+layer {
+  name: "roi_pool_conv5"
+  type: "ROIPooling"
+  bottom: "conv5"
+  bottom: "rois"
+  top: "roi_pool_conv5"
+  roi_pooling_param {
+    pooled_w: 6
+    pooled_h: 6
+    spatial_scale: 0.0625 # 1/16
+  }
+}
+layer {
+  name: "fc6"
+  type: "InnerProduct"
+  bottom: "roi_pool_conv5"
+  top: "fc6"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu6"
+  type: "ReLU"
+  bottom: "fc6"
+  top: "fc6"
+}
+layer {
+  name: "drop6"
+  type: "Dropout"
+  bottom: "fc6"
+  top: "fc6"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "fc7"
+  type: "InnerProduct"
+  bottom: "fc6"
+  top: "fc7"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 4096
+  }
+}
+layer {
+  name: "relu7"
+  type: "ReLU"
+  bottom: "fc7"
+  top: "fc7"
+}
+layer {
+  name: "drop7"
+  type: "Dropout"
+  bottom: "fc7"
+  top: "fc7"
+  dropout_param {
+    dropout_ratio: 0.5
+    scale_train: false
+  }
+}
+layer {
+  name: "cls_score"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "cls_score"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 21
+    weight_filler {
+      type: "gaussian"
+      std: 0.01
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "bbox_pred"
+  type: "InnerProduct"
+  bottom: "fc7"
+  top: "bbox_pred"
+  param { lr_mult: 1.0 }
+  param { lr_mult: 2.0 }
+  inner_product_param {
+    num_output: 84
+    weight_filler {
+      type: "gaussian"
+      std: 0.001
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layer {
+  name: "loss_cls"
+  type: "SoftmaxWithLoss"
+  bottom: "cls_score"
+  bottom: "labels"
+  propagate_down: 1
+  propagate_down: 0
+  top: "cls_loss"
+  loss_weight: 1
+  loss_param {
+    ignore_label: -1
+    normalize: true
+  }
+}
+layer {
+  name: "loss_bbox"
+  type: "SmoothL1Loss"
+  bottom: "bbox_pred"
+  bottom: "bbox_targets"
+  bottom: 'bbox_inside_weights'
+  bottom: 'bbox_outside_weights'
+  top: "bbox_loss"
+  loss_weight: 1
+}
diff --git a/tools/README.md b/tools/README.md
new file mode 100644
index 0000000..6fd4094
--- /dev/null
+++ b/tools/README.md
@@ -0,0 +1 @@
+Tools for training, testing, and compressing Fast R-CNN networks.
diff --git a/tools/_init_paths.py b/tools/_init_paths.py
new file mode 100644
index 0000000..f12404c
--- /dev/null
+++ b/tools/_init_paths.py
@@ -0,0 +1,25 @@
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Set up paths for Fast R-CNN."""
+
+import os.path as osp
+import sys
+
+def add_path(path):
+    if path not in sys.path:
+        sys.path.insert(0, path)
+
+this_dir = osp.dirname(__file__)
+
+# Add caffe to PYTHONPATH
+caffe_path = osp.join(this_dir, '..', 'caffe', 'python')
+add_path(caffe_path)
+
+# Add lib to PYTHONPATH
+lib_path = osp.join(this_dir, '..', 'lib')
+add_path(lib_path)
diff --git a/tools/compress_net.py b/tools/compress_net.py
new file mode 100755
index 0000000..e044e5b
--- /dev/null
+++ b/tools/compress_net.py
@@ -0,0 +1,125 @@
+#!/usr/bin/env python
+
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Compress a Fast R-CNN network using truncated SVD."""
+
+import _init_paths
+import caffe
+import argparse
+import numpy as np
+import os, sys
+
+def parse_args():
+    """Parse input arguments."""
+    parser = argparse.ArgumentParser(description='Compress a Fast R-CNN network')
+    parser.add_argument('--def', dest='prototxt',
+                        help='prototxt file defining the uncompressed network',
+                        default=None, type=str)
+    parser.add_argument('--def-svd', dest='prototxt_svd',
+                        help='prototxt file defining the SVD compressed network',
+                        default=None, type=str)
+    parser.add_argument('--net', dest='caffemodel',
+                        help='model to compress',
+                        default=None, type=str)
+
+    if len(sys.argv) == 1:
+        parser.print_help()
+        sys.exit(1)
+
+    args = parser.parse_args()
+    return args
+
+def compress_weights(W, l):
+    """Compress the weight matrix W of an inner product (fully connected) layer
+    using truncated SVD.
+
+    Parameters:
+    W: N x M weights matrix
+    l: number of singular values to retain
+
+    Returns:
+    Ul, L: matrices such that W \approx Ul*L
+    """
+
+    # numpy doesn't seem to have a fast truncated SVD algorithm...
+    # this could be faster
+    U, s, V = np.linalg.svd(W, full_matrices=False)
+
+    Ul = U[:, :l]
+    sl = s[:l]
+    Vl = V[:l, :]
+
+    L = np.dot(np.diag(sl), Vl)
+    return Ul, L
+
+def main():
+    args = parse_args()
+
+    # prototxt = 'models/VGG16/test.prototxt'
+    # caffemodel = 'snapshots/vgg16_fast_rcnn_iter_40000.caffemodel'
+    net = caffe.Net(args.prototxt, args.caffemodel, caffe.TEST)
+
+    # prototxt_svd = 'models/VGG16/svd/test_fc6_fc7.prototxt'
+    # caffemodel = 'snapshots/vgg16_fast_rcnn_iter_40000.caffemodel'
+    net_svd = caffe.Net(args.prototxt_svd, args.caffemodel, caffe.TEST)
+
+    print('Uncompressed network {} : {}'.format(args.prototxt, args.caffemodel))
+    print('Compressed network prototxt {}'.format(args.prototxt_svd))
+
+    out = os.path.splitext(os.path.basename(args.caffemodel))[0] + '_svd'
+    out_dir = os.path.dirname(args.caffemodel)
+
+    # Compress fc6
+    if net_svd.params.has_key('fc6_L'):
+        l_fc6 = net_svd.params['fc6_L'][0].data.shape[0]
+        print('  fc6_L bottleneck size: {}'.format(l_fc6))
+
+        # uncompressed weights and biases
+        W_fc6 = net.params['fc6'][0].data
+        B_fc6 = net.params['fc6'][1].data
+
+        print('  compressing fc6...')
+        Ul_fc6, L_fc6 = compress_weights(W_fc6, l_fc6)
+
+        assert(len(net_svd.params['fc6_L']) == 1)
+
+        # install compressed matrix factors (and original biases)
+        net_svd.params['fc6_L'][0].data[...] = L_fc6
+
+        net_svd.params['fc6_U'][0].data[...] = Ul_fc6
+        net_svd.params['fc6_U'][1].data[...] = B_fc6
+
+        out += '_fc6_{}'.format(l_fc6)
+
+    # Compress fc7
+    if net_svd.params.has_key('fc7_L'):
+        l_fc7 = net_svd.params['fc7_L'][0].data.shape[0]
+        print '  fc7_L bottleneck size: {}'.format(l_fc7)
+
+        W_fc7 = net.params['fc7'][0].data
+        B_fc7 = net.params['fc7'][1].data
+
+        print('  compressing fc7...')
+        Ul_fc7, L_fc7 = compress_weights(W_fc7, l_fc7)
+
+        assert(len(net_svd.params['fc7_L']) == 1)
+
+        net_svd.params['fc7_L'][0].data[...] = L_fc7
+
+        net_svd.params['fc7_U'][0].data[...] = Ul_fc7
+        net_svd.params['fc7_U'][1].data[...] = B_fc7
+
+        out += '_fc7_{}'.format(l_fc7)
+
+    filename = '{}/{}.caffemodel'.format(out_dir, out)
+    net_svd.save(filename)
+    print 'Wrote svd model to: {:s}'.format(filename)
+
+if __name__ == '__main__':
+    main()
diff --git a/tools/demo.py b/tools/demo.py
new file mode 100755
index 0000000..631c68a
--- /dev/null
+++ b/tools/demo.py
@@ -0,0 +1,151 @@
+#!/usr/bin/env python
+
+# --------------------------------------------------------
+# Faster R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""
+Demo script showing detections in sample images.
+
+See README.md for installation instructions before running.
+"""
+
+import _init_paths
+from fast_rcnn.config import cfg
+from fast_rcnn.test import im_detect
+from fast_rcnn.nms_wrapper import nms
+from utils.timer import Timer
+import matplotlib.pyplot as plt
+import numpy as np
+import scipy.io as sio
+import caffe, os, sys, cv2
+import argparse
+
+CLASSES = ('__background__',
+           'aeroplane', 'bicycle', 'bird', 'boat',
+           'bottle', 'bus', 'car', 'cat', 'chair',
+           'cow', 'diningtable', 'dog', 'horse',
+           'motorbike', 'person', 'pottedplant',
+           'sheep', 'sofa', 'train', 'tvmonitor')
+
+NETS = {'vgg16': ('VGG16',
+                  'VGG16_faster_rcnn_final.caffemodel'),
+        'zf': ('ZF',
+                  'ZF_faster_rcnn_final.caffemodel')}
+
+
+def vis_detections(im, class_name, dets, thresh=0.5):
+    """Draw detected bounding boxes."""
+    inds = np.where(dets[:, -1] >= thresh)[0]
+    if len(inds) == 0:
+        return
+
+    im = im[:, :, (2, 1, 0)]
+    fig, ax = plt.subplots(figsize=(12, 12))
+    ax.imshow(im, aspect='equal')
+    for i in inds:
+        bbox = dets[i, :4]
+        score = dets[i, -1]
+
+        ax.add_patch(
+            plt.Rectangle((bbox[0], bbox[1]),
+                          bbox[2] - bbox[0],
+                          bbox[3] - bbox[1], fill=False,
+                          edgecolor='red', linewidth=3.5)
+            )
+        ax.text(bbox[0], bbox[1] - 2,
+                '{:s} {:.3f}'.format(class_name, score),
+                bbox=dict(facecolor='blue', alpha=0.5),
+                fontsize=14, color='white')
+
+    ax.set_title(('{} detections with '
+                  'p({} | box) >= {:.1f}').format(class_name, class_name,
+                                                  thresh),
+                  fontsize=14)
+    plt.axis('off')
+    plt.tight_layout()
+    plt.draw()
+
+def demo(net, image_name):
+    """Detect object classes in an image using pre-computed object proposals."""
+
+    # Load the demo image
+    im_file = os.path.join(cfg.DATA_DIR, 'demo', image_name)
+    im = cv2.imread(im_file)
+
+    # Detect all object classes and regress object bounds
+    timer = Timer()
+    timer.tic()
+    scores, boxes = im_detect(net, im)
+    timer.toc()
+    print ('Detection took {:.3f}s for '
+           '{:d} object proposals').format(timer.total_time, boxes.shape[0])
+
+    # Visualize detections for each class
+    CONF_THRESH = 0.8
+    NMS_THRESH = 0.3
+    for cls_ind, cls in enumerate(CLASSES[1:]):
+        cls_ind += 1 # because we skipped background
+        cls_boxes = boxes[:, 4*cls_ind:4*(cls_ind + 1)]
+        cls_scores = scores[:, cls_ind]
+        dets = np.hstack((cls_boxes,
+                          cls_scores[:, np.newaxis])).astype(np.float32)
+        keep = nms(dets, NMS_THRESH)
+        dets = dets[keep, :]
+        vis_detections(im, cls, dets, thresh=CONF_THRESH)
+
+def parse_args():
+    """Parse input arguments."""
+    parser = argparse.ArgumentParser(description='Faster R-CNN demo')
+    parser.add_argument('--gpu', dest='gpu_id', help='GPU device id to use [0]',
+                        default=0, type=int)
+    parser.add_argument('--cpu', dest='cpu_mode',
+                        help='Use CPU mode (overrides --gpu)',
+                        action='store_true')
+    parser.add_argument('--net', dest='demo_net', help='Network to use [vgg16]',
+                        choices=NETS.keys(), default='vgg16')
+
+    args = parser.parse_args()
+
+    return args
+
+if __name__ == '__main__':
+    cfg.TEST.HAS_RPN = True  # Use RPN for proposals
+
+    args = parse_args()
+
+    prototxt = os.path.join(cfg.MODELS_DIR, NETS[args.demo_net][0],
+                            'faster_rcnn_alt_opt', 'faster_rcnn_test.pt')
+    caffemodel = os.path.join(cfg.DATA_DIR, 'faster_rcnn_models',
+                              NETS[args.demo_net][1])
+
+    if not os.path.isfile(caffemodel):
+        raise IOError(('{:s} not found.\nDid you run ./data/script/'
+                       'fetch_faster_rcnn_models.sh?').format(caffemodel))
+
+    if args.cpu_mode:
+        caffe.set_mode_cpu()
+    else:
+        caffe.set_mode_gpu()
+        caffe.set_device(args.gpu_id)
+        cfg.GPU_ID = args.gpu_id
+    net = caffe.Net(prototxt, caffemodel, caffe.TEST)
+
+    print '\n\nLoaded network {:s}'.format(caffemodel)
+
+    # Warmup on a dummy image
+    im = 128 * np.ones((300, 500, 3), dtype=np.uint8)
+    for i in xrange(2):
+        _, _= im_detect(net, im)
+
+    im_names = ['000456.jpg', '000542.jpg', '001150.jpg',
+                '001763.jpg', '004545.jpg']
+    for im_name in im_names:
+        print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+        print 'Demo for data/demo/{}'.format(im_name)
+        demo(net, im_name)
+
+    plt.show()
diff --git a/tools/eval_recall.py b/tools/eval_recall.py
new file mode 100755
index 0000000..b1a59dc
--- /dev/null
+++ b/tools/eval_recall.py
@@ -0,0 +1,70 @@
+#!/usr/bin/env python
+
+import _init_paths
+from fast_rcnn.config import cfg, cfg_from_file, cfg_from_list
+from datasets.factory import get_imdb
+import argparse
+import time, os, sys
+import numpy as np
+
+def parse_args():
+    """
+    Parse input arguments
+    """
+    parser = argparse.ArgumentParser(description='Test a Fast R-CNN network')
+    parser.add_argument('--imdb', dest='imdb_name',
+                        help='dataset to test',
+                        default='voc_2007_test', type=str)
+    parser.add_argument('--method', dest='method',
+                        help='proposal method',
+                        default='selective_search', type=str)
+    parser.add_argument('--rpn-file', dest='rpn_file',
+                        default=None, type=str)
+
+    if len(sys.argv) == 1:
+        parser.print_help()
+        sys.exit(1)
+
+    args = parser.parse_args()
+    return args
+
+if __name__ == '__main__':
+    args = parse_args()
+
+    print('Called with args:')
+    print(args)
+
+    imdb = get_imdb(args.imdb_name)
+    imdb.set_proposal_method(args.method)
+    if args.rpn_file is not None:
+        imdb.config['rpn_file'] = args.rpn_file
+
+    candidate_boxes = None
+    if 0:
+        import scipy.io as sio
+        filename = 'debug/stage1_rpn_voc_2007_test.mat'
+        raw_data = sio.loadmat(filename)['aboxes'].ravel()
+        candidate_boxes = raw_data
+
+    ar, gt_overlaps, recalls, thresholds = \
+        imdb.evaluate_recall(candidate_boxes=candidate_boxes)
+    print 'Method: {}'.format(args.method)
+    print 'AverageRec: {:.3f}'.format(ar)
+
+    def recall_at(t):
+        ind = np.where(thresholds > t - 1e-5)[0][0]
+        assert np.isclose(thresholds[ind], t)
+        return recalls[ind]
+
+    print 'Recall@0.5: {:.3f}'.format(recall_at(0.5))
+    print 'Recall@0.6: {:.3f}'.format(recall_at(0.6))
+    print 'Recall@0.7: {:.3f}'.format(recall_at(0.7))
+    print 'Recall@0.8: {:.3f}'.format(recall_at(0.8))
+    print 'Recall@0.9: {:.3f}'.format(recall_at(0.9))
+    # print again for easy spreadsheet copying
+    print '{:.3f}'.format(ar)
+    print '{:.3f}'.format(recall_at(0.5))
+    print '{:.3f}'.format(recall_at(0.6))
+    print '{:.3f}'.format(recall_at(0.7))
+    print '{:.3f}'.format(recall_at(0.8))
+    print '{:.3f}'.format(recall_at(0.9))
diff --git a/tools/reval.py b/tools/reval.py
new file mode 100755
index 0000000..905ec1b
--- /dev/null
+++ b/tools/reval.py
@@ -0,0 +1,66 @@
+#!/usr/bin/env python
+
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Reval = re-eval. Re-evaluate saved detections."""
+
+import _init_paths
+from fast_rcnn.test import apply_nms
+from fast_rcnn.config import cfg
+from datasets.factory import get_imdb
+import cPickle
+import os, sys, argparse
+import numpy as np
+
+def parse_args():
+    """
+    Parse input arguments
+    """
+    parser = argparse.ArgumentParser(description='Re-evaluate results')
+    parser.add_argument('output_dir', nargs=1, help='results directory',
+                        type=str)
+    parser.add_argument('--imdb', dest='imdb_name',
+                        help='dataset to re-evaluate',
+                        default='voc_2007_test', type=str)
+    parser.add_argument('--matlab', dest='matlab_eval',
+                        help='use matlab for evaluation',
+                        action='store_true')
+    parser.add_argument('--comp', dest='comp_mode', help='competition mode',
+                        action='store_true')
+    parser.add_argument('--nms', dest='apply_nms', help='apply nms',
+                        action='store_true')
+
+    if len(sys.argv) == 1:
+        parser.print_help()
+        sys.exit(1)
+
+    args = parser.parse_args()
+    return args
+
+def from_dets(imdb_name, output_dir, args):
+    imdb = get_imdb(imdb_name)
+    imdb.competition_mode(args.comp_mode)
+    imdb.config['matlab_eval'] = args.matlab_eval
+    with open(os.path.join(output_dir, 'detections.pkl'), 'rb') as f:
+        dets = cPickle.load(f)
+
+    if args.apply_nms:
+        print 'Applying NMS to all detections'
+        nms_dets = apply_nms(dets, cfg.TEST.NMS)
+    else:
+        nms_dets = dets
+
+    print 'Evaluating detections'
+    imdb.evaluate_detections(nms_dets, output_dir)
+
+if __name__ == '__main__':
+    args = parse_args()
+
+    output_dir = os.path.abspath(args.output_dir[0])
+    imdb_name = args.imdb_name
+    from_dets(imdb_name, output_dir, args)
diff --git a/tools/rpn_generate.py b/tools/rpn_generate.py
new file mode 100755
index 0000000..f8ca4a1
--- /dev/null
+++ b/tools/rpn_generate.py
@@ -0,0 +1,91 @@
+#!/usr/bin/env python
+
+# --------------------------------------------------------
+# Fast/er/ R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Generate RPN proposals."""
+
+import _init_paths
+import numpy as np
+from fast_rcnn.config import cfg, cfg_from_file, cfg_from_list, get_output_dir
+from datasets.factory import get_imdb
+from rpn.generate import imdb_proposals
+import cPickle
+import caffe
+import argparse
+import pprint
+import time, os, sys
+
+def parse_args():
+    """
+    Parse input arguments
+    """
+    parser = argparse.ArgumentParser(description='Test a Fast R-CNN network')
+    parser.add_argument('--gpu', dest='gpu_id', help='GPU id to use',
+                        default=0, type=int)
+    parser.add_argument('--def', dest='prototxt',
+                        help='prototxt file defining the network',
+                        default=None, type=str)
+    parser.add_argument('--net', dest='caffemodel',
+                        help='model to test',
+                        default=None, type=str)
+    parser.add_argument('--cfg', dest='cfg_file',
+                        help='optional config file', default=None, type=str)
+    parser.add_argument('--wait', dest='wait',
+                        help='wait until net file exists',
+                        default=True, type=bool)
+    parser.add_argument('--imdb', dest='imdb_name',
+                        help='dataset to test',
+                        default='voc_2007_test', type=str)
+    parser.add_argument('--set', dest='set_cfgs',
+                        help='set config keys', default=None,
+                        nargs=argparse.REMAINDER)
+
+    if len(sys.argv) == 1:
+        parser.print_help()
+        sys.exit(1)
+
+    args = parser.parse_args()
+    return args
+
+if __name__ == '__main__':
+    args = parse_args()
+
+    print('Called with args:')
+    print(args)
+
+    if args.cfg_file is not None:
+        cfg_from_file(args.cfg_file)
+    if args.set_cfgs is not None:
+        cfg_from_list(args.set_cfgs)
+
+    cfg.GPU_ID = args.gpu_id
+
+    # RPN test settings
+    cfg.TEST.RPN_PRE_NMS_TOP_N = -1
+    cfg.TEST.RPN_POST_NMS_TOP_N = 2000
+
+    print('Using config:')
+    pprint.pprint(cfg)
+
+    while not os.path.exists(args.caffemodel) and args.wait:
+        print('Waiting for {} to exist...'.format(args.caffemodel))
+        time.sleep(10)
+
+    caffe.set_mode_gpu()
+    caffe.set_device(args.gpu_id)
+    net = caffe.Net(args.prototxt, args.caffemodel, caffe.TEST)
+    net.name = os.path.splitext(os.path.basename(args.caffemodel))[0]
+
+    imdb = get_imdb(args.imdb_name)
+    imdb_boxes = imdb_proposals(net, imdb)
+
+    output_dir = get_output_dir(imdb, net)
+    rpn_file = os.path.join(output_dir, net.name + '_rpn_proposals.pkl')
+    with open(rpn_file, 'wb') as f:
+        cPickle.dump(imdb_boxes, f, cPickle.HIGHEST_PROTOCOL)
+    print 'Wrote RPN proposals to {}'.format(rpn_file)
diff --git a/tools/test_net.py b/tools/test_net.py
new file mode 100755
index 0000000..de4f12b
--- /dev/null
+++ b/tools/test_net.py
@@ -0,0 +1,90 @@
+#!/usr/bin/env python
+
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Test a Fast R-CNN network on an image database."""
+
+import _init_paths
+from fast_rcnn.test import test_net
+from fast_rcnn.config import cfg, cfg_from_file, cfg_from_list
+from datasets.factory import get_imdb
+import caffe
+import argparse
+import pprint
+import time, os, sys
+
+def parse_args():
+    """
+    Parse input arguments
+    """
+    parser = argparse.ArgumentParser(description='Test a Fast R-CNN network')
+    parser.add_argument('--gpu', dest='gpu_id', help='GPU id to use',
+                        default=0, type=int)
+    parser.add_argument('--def', dest='prototxt',
+                        help='prototxt file defining the network',
+                        default=None, type=str)
+    parser.add_argument('--net', dest='caffemodel',
+                        help='model to test',
+                        default=None, type=str)
+    parser.add_argument('--cfg', dest='cfg_file',
+                        help='optional config file', default=None, type=str)
+    parser.add_argument('--wait', dest='wait',
+                        help='wait until net file exists',
+                        default=True, type=bool)
+    parser.add_argument('--imdb', dest='imdb_name',
+                        help='dataset to test',
+                        default='voc_2007_test', type=str)
+    parser.add_argument('--comp', dest='comp_mode', help='competition mode',
+                        action='store_true')
+    parser.add_argument('--set', dest='set_cfgs',
+                        help='set config keys', default=None,
+                        nargs=argparse.REMAINDER)
+    parser.add_argument('--vis', dest='vis', help='visualize detections',
+                        action='store_true')
+    parser.add_argument('--num_dets', dest='max_per_image',
+                        help='max number of detections per image',
+                        default=100, type=int)
+
+    if len(sys.argv) == 1:
+        parser.print_help()
+        sys.exit(1)
+
+    args = parser.parse_args()
+    return args
+
+if __name__ == '__main__':
+    args = parse_args()
+
+    print('Called with args:')
+    print(args)
+
+    if args.cfg_file is not None:
+        cfg_from_file(args.cfg_file)
+    if args.set_cfgs is not None:
+        cfg_from_list(args.set_cfgs)
+
+    cfg.GPU_ID = args.gpu_id
+
+    print('Using config:')
+    pprint.pprint(cfg)
+
+    while not os.path.exists(args.caffemodel) and args.wait:
+        print('Waiting for {} to exist...'.format(args.caffemodel))
+        time.sleep(10)
+
+    caffe.set_mode_gpu()
+    caffe.set_device(args.gpu_id)
+    net = caffe.Net(args.prototxt, args.caffemodel, caffe.TEST)
+    net.name = os.path.splitext(os.path.basename(args.caffemodel))[0]
+
+    imdb = get_imdb(args.imdb_name)
+    imdb.competition_mode(args.comp_mode)
+    if not cfg.TEST.HAS_RPN:
+        imdb.set_proposal_method(cfg.TEST.PROPOSAL_METHOD)
+
+    test_net(net, imdb, max_per_image=args.max_per_image, vis=args.vis)
diff --git a/tools/train_faster_rcnn_alt_opt.py b/tools/train_faster_rcnn_alt_opt.py
new file mode 100755
index 0000000..e49844a
--- /dev/null
+++ b/tools/train_faster_rcnn_alt_opt.py
@@ -0,0 +1,334 @@
+#!/usr/bin/env python
+
+# --------------------------------------------------------
+# Faster R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Train a Faster R-CNN network using alternating optimization.
+This tool implements the alternating optimization algorithm described in our
+NIPS 2015 paper ("Faster R-CNN: Towards Real-time Object Detection with Region
+Proposal Networks." Shaoqing Ren, Kaiming He, Ross Girshick, Jian Sun.)
+"""
+
+import _init_paths
+from fast_rcnn.train import get_training_roidb, train_net
+from fast_rcnn.config import cfg, cfg_from_file, cfg_from_list, get_output_dir
+from datasets.factory import get_imdb
+from rpn.generate import imdb_proposals
+import argparse
+import pprint
+import numpy as np
+import sys, os
+import multiprocessing as mp
+import cPickle
+import shutil
+
+def parse_args():
+    """
+    Parse input arguments
+    """
+    parser = argparse.ArgumentParser(description='Train a Faster R-CNN network')
+    parser.add_argument('--gpu', dest='gpu_id',
+                        help='GPU device id to use [0]',
+                        default=0, type=int)
+    parser.add_argument('--net_name', dest='net_name',
+                        help='network name (e.g., "ZF")',
+                        default=None, type=str)
+    parser.add_argument('--weights', dest='pretrained_model',
+                        help='initialize with pretrained model weights',
+                        default=None, type=str)
+    parser.add_argument('--cfg', dest='cfg_file',
+                        help='optional config file',
+                        default=None, type=str)
+    parser.add_argument('--imdb', dest='imdb_name',
+                        help='dataset to train on',
+                        default='voc_2007_trainval', type=str)
+    parser.add_argument('--set', dest='set_cfgs',
+                        help='set config keys', default=None,
+                        nargs=argparse.REMAINDER)
+
+    if len(sys.argv) == 1:
+        parser.print_help()
+        sys.exit(1)
+
+    args = parser.parse_args()
+    return args
+
+def get_roidb(imdb_name, rpn_file=None):
+    imdb = get_imdb(imdb_name)
+    print 'Loaded dataset `{:s}` for training'.format(imdb.name)
+    imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD)
+    print 'Set proposal method: {:s}'.format(cfg.TRAIN.PROPOSAL_METHOD)
+    if rpn_file is not None:
+        imdb.config['rpn_file'] = rpn_file
+    roidb = get_training_roidb(imdb)
+    return roidb, imdb
+
+def get_solvers(net_name):
+    # Faster R-CNN Alternating Optimization
+    n = 'faster_rcnn_alt_opt'
+    # Solver for each training stage
+    solvers = [[net_name, n, 'stage1_rpn_solver60k80k.pt'],
+               [net_name, n, 'stage1_fast_rcnn_solver30k40k.pt'],
+               [net_name, n, 'stage2_rpn_solver60k80k.pt'],
+               [net_name, n, 'stage2_fast_rcnn_solver30k40k.pt']]
+    solvers = [os.path.join(cfg.MODELS_DIR, *s) for s in solvers]
+    # Iterations for each training stage
+    max_iters = [80000, 40000, 80000, 40000]
+    # max_iters = [100, 100, 100, 100]
+    # Test prototxt for the RPN
+    rpn_test_prototxt = os.path.join(
+        cfg.MODELS_DIR, net_name, n, 'rpn_test.pt')
+    return solvers, max_iters, rpn_test_prototxt
+
+# ------------------------------------------------------------------------------
+# Pycaffe doesn't reliably free GPU memory when instantiated nets are discarded
+# (e.g. "del net" in Python code). To work around this issue, each training
+# stage is executed in a separate process using multiprocessing.Process.
+# ------------------------------------------------------------------------------
+
+def _init_caffe(cfg):
+    """Initialize pycaffe in a training process.
+    """
+
+    import caffe
+    # fix the random seeds (numpy and caffe) for reproducibility
+    np.random.seed(cfg.RNG_SEED)
+    caffe.set_random_seed(cfg.RNG_SEED)
+    # set up caffe
+    caffe.set_mode_gpu()
+    caffe.set_device(cfg.GPU_ID)
+
+def train_rpn(queue=None, imdb_name=None, init_model=None, solver=None,
+              max_iters=None, cfg=None):
+    """Train a Region Proposal Network in a separate training process.
+    """
+
+    # Not using any proposals, just ground-truth boxes
+    cfg.TRAIN.HAS_RPN = True
+    cfg.TRAIN.BBOX_REG = False  # applies only to Fast R-CNN bbox regression
+    cfg.TRAIN.PROPOSAL_METHOD = 'gt'
+    cfg.TRAIN.IMS_PER_BATCH = 1
+    print 'Init model: {}'.format(init_model)
+    print('Using config:')
+    pprint.pprint(cfg)
+
+    import caffe
+    _init_caffe(cfg)
+
+    roidb, imdb = get_roidb(imdb_name)
+    print 'roidb len: {}'.format(len(roidb))
+    output_dir = get_output_dir(imdb)
+    print 'Output will be saved to `{:s}`'.format(output_dir)
+
+    model_paths = train_net(solver, roidb, output_dir,
+                            pretrained_model=init_model,
+                            max_iters=max_iters)
+    # Cleanup all but the final model
+    for i in model_paths[:-1]:
+        os.remove(i)
+    rpn_model_path = model_paths[-1]
+    # Send final model path through the multiprocessing queue
+    queue.put({'model_path': rpn_model_path})
+
+def rpn_generate(queue=None, imdb_name=None, rpn_model_path=None, cfg=None,
+                 rpn_test_prototxt=None):
+    """Use a trained RPN to generate proposals.
+    """
+
+    cfg.TEST.RPN_PRE_NMS_TOP_N = -1     # no pre NMS filtering
+    cfg.TEST.RPN_POST_NMS_TOP_N = 2000  # limit top boxes after NMS
+    print 'RPN model: {}'.format(rpn_model_path)
+    print('Using config:')
+    pprint.pprint(cfg)
+
+    import caffe
+    _init_caffe(cfg)
+
+    # NOTE: the matlab implementation computes proposals on flipped images, too.
+    # We compute them on the image once and then flip the already computed
+    # proposals. This might cause a minor loss in mAP (less proposal jittering).
+    imdb = get_imdb(imdb_name)
+    print 'Loaded dataset `{:s}` for proposal generation'.format(imdb.name)
+
+    # Load RPN and configure output directory
+    rpn_net = caffe.Net(rpn_test_prototxt, rpn_model_path, caffe.TEST)
+    output_dir = get_output_dir(imdb)
+    print 'Output will be saved to `{:s}`'.format(output_dir)
+    # Generate proposals on the imdb
+    rpn_proposals = imdb_proposals(rpn_net, imdb)
+    # Write proposals to disk and send the proposal file path through the
+    # multiprocessing queue
+    rpn_net_name = os.path.splitext(os.path.basename(rpn_model_path))[0]
+    rpn_proposals_path = os.path.join(
+        output_dir, rpn_net_name + '_proposals.pkl')
+    with open(rpn_proposals_path, 'wb') as f:
+        cPickle.dump(rpn_proposals, f, cPickle.HIGHEST_PROTOCOL)
+    print 'Wrote RPN proposals to {}'.format(rpn_proposals_path)
+    queue.put({'proposal_path': rpn_proposals_path})
+
+def train_fast_rcnn(queue=None, imdb_name=None, init_model=None, solver=None,
+                    max_iters=None, cfg=None, rpn_file=None):
+    """Train a Fast R-CNN using proposals generated by an RPN.
+    """
+
+    cfg.TRAIN.HAS_RPN = False           # not generating prosals on-the-fly
+    cfg.TRAIN.PROPOSAL_METHOD = 'rpn'   # use pre-computed RPN proposals instead
+    cfg.TRAIN.IMS_PER_BATCH = 2
+    print 'Init model: {}'.format(init_model)
+    print 'RPN proposals: {}'.format(rpn_file)
+    print('Using config:')
+    pprint.pprint(cfg)
+
+    import caffe
+    _init_caffe(cfg)
+
+    roidb, imdb = get_roidb(imdb_name, rpn_file=rpn_file)
+    output_dir = get_output_dir(imdb)
+    print 'Output will be saved to `{:s}`'.format(output_dir)
+    # Train Fast R-CNN
+    model_paths = train_net(solver, roidb, output_dir,
+                            pretrained_model=init_model,
+                            max_iters=max_iters)
+    # Cleanup all but the final model
+    for i in model_paths[:-1]:
+        os.remove(i)
+    fast_rcnn_model_path = model_paths[-1]
+    # Send Fast R-CNN model path over the multiprocessing queue
+    queue.put({'model_path': fast_rcnn_model_path})
+
+if __name__ == '__main__':
+    args = parse_args()
+
+    print('Called with args:')
+    print(args)
+
+    if args.cfg_file is not None:
+        cfg_from_file(args.cfg_file)
+    if args.set_cfgs is not None:
+        cfg_from_list(args.set_cfgs)
+    cfg.GPU_ID = args.gpu_id
+
+    # --------------------------------------------------------------------------
+    # Pycaffe doesn't reliably free GPU memory when instantiated nets are
+    # discarded (e.g. "del net" in Python code). To work around this issue, each
+    # training stage is executed in a separate process using
+    # multiprocessing.Process.
+    # --------------------------------------------------------------------------
+
+    # queue for communicated results between processes
+    mp_queue = mp.Queue()
+    # solves, iters, etc. for each training stage
+    solvers, max_iters, rpn_test_prototxt = get_solvers(args.net_name)
+
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+    print 'Stage 1 RPN, init from ImageNet model'
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+
+    cfg.TRAIN.SNAPSHOT_INFIX = 'stage1'
+    mp_kwargs = dict(
+            queue=mp_queue,
+            imdb_name=args.imdb_name,
+            init_model=args.pretrained_model,
+            solver=solvers[0],
+            max_iters=max_iters[0],
+            cfg=cfg)
+    p = mp.Process(target=train_rpn, kwargs=mp_kwargs)
+    p.start()
+    rpn_stage1_out = mp_queue.get()
+    p.join()
+
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+    print 'Stage 1 RPN, generate proposals'
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+
+    mp_kwargs = dict(
+            queue=mp_queue,
+            imdb_name=args.imdb_name,
+            rpn_model_path=str(rpn_stage1_out['model_path']),
+            cfg=cfg,
+            rpn_test_prototxt=rpn_test_prototxt)
+    p = mp.Process(target=rpn_generate, kwargs=mp_kwargs)
+    p.start()
+    rpn_stage1_out['proposal_path'] = mp_queue.get()['proposal_path']
+    p.join()
+
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+    print 'Stage 1 Fast R-CNN using RPN proposals, init from ImageNet model'
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+
+    cfg.TRAIN.SNAPSHOT_INFIX = 'stage1'
+    mp_kwargs = dict(
+            queue=mp_queue,
+            imdb_name=args.imdb_name,
+            init_model=args.pretrained_model,
+            solver=solvers[1],
+            max_iters=max_iters[1],
+            cfg=cfg,
+            rpn_file=rpn_stage1_out['proposal_path'])
+    p = mp.Process(target=train_fast_rcnn, kwargs=mp_kwargs)
+    p.start()
+    fast_rcnn_stage1_out = mp_queue.get()
+    p.join()
+
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+    print 'Stage 2 RPN, init from stage 1 Fast R-CNN model'
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+
+    cfg.TRAIN.SNAPSHOT_INFIX = 'stage2'
+    mp_kwargs = dict(
+            queue=mp_queue,
+            imdb_name=args.imdb_name,
+            init_model=str(fast_rcnn_stage1_out['model_path']),
+            solver=solvers[2],
+            max_iters=max_iters[2],
+            cfg=cfg)
+    p = mp.Process(target=train_rpn, kwargs=mp_kwargs)
+    p.start()
+    rpn_stage2_out = mp_queue.get()
+    p.join()
+
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+    print 'Stage 2 RPN, generate proposals'
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+
+    mp_kwargs = dict(
+            queue=mp_queue,
+            imdb_name=args.imdb_name,
+            rpn_model_path=str(rpn_stage2_out['model_path']),
+            cfg=cfg,
+            rpn_test_prototxt=rpn_test_prototxt)
+    p = mp.Process(target=rpn_generate, kwargs=mp_kwargs)
+    p.start()
+    rpn_stage2_out['proposal_path'] = mp_queue.get()['proposal_path']
+    p.join()
+
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+    print 'Stage 2 Fast R-CNN, init from stage 2 RPN R-CNN model'
+    print '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
+
+    cfg.TRAIN.SNAPSHOT_INFIX = 'stage2'
+    mp_kwargs = dict(
+            queue=mp_queue,
+            imdb_name=args.imdb_name,
+            init_model=str(rpn_stage2_out['model_path']),
+            solver=solvers[3],
+            max_iters=max_iters[3],
+            cfg=cfg,
+            rpn_file=rpn_stage2_out['proposal_path'])
+    p = mp.Process(target=train_fast_rcnn, kwargs=mp_kwargs)
+    p.start()
+    fast_rcnn_stage2_out = mp_queue.get()
+    p.join()
+
+    # Create final model (just a copy of the last stage)
+    final_path = os.path.join(
+            os.path.dirname(fast_rcnn_stage2_out['model_path']),
+            args.net_name + '_faster_rcnn_final.caffemodel')
+    print 'cp {} -> {}'.format(
+            fast_rcnn_stage2_out['model_path'], final_path)
+    shutil.copy(fast_rcnn_stage2_out['model_path'], final_path)
+    print 'Final model: {}'.format(final_path)
diff --git a/tools/train_net.py b/tools/train_net.py
new file mode 100755
index 0000000..622a95d
--- /dev/null
+++ b/tools/train_net.py
@@ -0,0 +1,112 @@
+#!/usr/bin/env python
+
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""Train a Fast R-CNN network on a region of interest database."""
+
+import _init_paths
+from fast_rcnn.train import get_training_roidb, train_net
+from fast_rcnn.config import cfg, cfg_from_file, cfg_from_list, get_output_dir
+from datasets.factory import get_imdb
+import datasets.imdb
+import caffe
+import argparse
+import pprint
+import numpy as np
+import sys
+
+def parse_args():
+    """
+    Parse input arguments
+    """
+    parser = argparse.ArgumentParser(description='Train a Fast R-CNN network')
+    parser.add_argument('--gpu', dest='gpu_id',
+                        help='GPU device id to use [0]',
+                        default=0, type=int)
+    parser.add_argument('--solver', dest='solver',
+                        help='solver prototxt',
+                        default=None, type=str)
+    parser.add_argument('--iters', dest='max_iters',
+                        help='number of iterations to train',
+                        default=40000, type=int)
+    parser.add_argument('--weights', dest='pretrained_model',
+                        help='initialize with pretrained model weights',
+                        default=None, type=str)
+    parser.add_argument('--cfg', dest='cfg_file',
+                        help='optional config file',
+                        default=None, type=str)
+    parser.add_argument('--imdb', dest='imdb_name',
+                        help='dataset to train on',
+                        default='voc_2007_trainval', type=str)
+    parser.add_argument('--rand', dest='randomize',
+                        help='randomize (do not use a fixed seed)',
+                        action='store_true')
+    parser.add_argument('--set', dest='set_cfgs',
+                        help='set config keys', default=None,
+                        nargs=argparse.REMAINDER)
+
+    if len(sys.argv) == 1:
+        parser.print_help()
+        sys.exit(1)
+
+    args = parser.parse_args()
+    return args
+
+def combined_roidb(imdb_names):
+    def get_roidb(imdb_name):
+        imdb = get_imdb(imdb_name)
+        print 'Loaded dataset `{:s}` for training'.format(imdb.name)
+        imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD)
+        print 'Set proposal method: {:s}'.format(cfg.TRAIN.PROPOSAL_METHOD)
+        roidb = get_training_roidb(imdb)
+        return roidb
+
+    roidbs = [get_roidb(s) for s in imdb_names.split('+')]
+    roidb = roidbs[0]
+    if len(roidbs) > 1:
+        for r in roidbs[1:]:
+            roidb.extend(r)
+        imdb = datasets.imdb.imdb(imdb_names)
+    else:
+        imdb = get_imdb(imdb_names)
+    return imdb, roidb
+
+if __name__ == '__main__':
+    args = parse_args()
+
+    print('Called with args:')
+    print(args)
+
+    if args.cfg_file is not None:
+        cfg_from_file(args.cfg_file)
+    if args.set_cfgs is not None:
+        cfg_from_list(args.set_cfgs)
+
+    cfg.GPU_ID = args.gpu_id
+
+    print('Using config:')
+    pprint.pprint(cfg)
+
+    if not args.randomize:
+        # fix the random seeds (numpy and caffe) for reproducibility
+        np.random.seed(cfg.RNG_SEED)
+        caffe.set_random_seed(cfg.RNG_SEED)
+
+    # set up caffe
+    caffe.set_mode_gpu()
+    caffe.set_device(args.gpu_id)
+
+    imdb, roidb = combined_roidb(args.imdb_name)
+    print '{:d} roidb entries'.format(len(roidb))
+
+    output_dir = get_output_dir(imdb)
+    print 'Output will be saved to `{:s}`'.format(output_dir)
+
+    train_net(args.solver, roidb, output_dir,
+              pretrained_model=args.pretrained_model,
+              max_iters=args.max_iters)
diff --git a/tools/train_svms.py b/tools/train_svms.py
new file mode 100755
index 0000000..498bbf2
--- /dev/null
+++ b/tools/train_svms.py
@@ -0,0 +1,353 @@
+#!/usr/bin/env python
+
+# --------------------------------------------------------
+# Fast R-CNN
+# Copyright (c) 2015 Microsoft
+# Licensed under The MIT License [see LICENSE for details]
+# Written by Ross Girshick
+# --------------------------------------------------------
+
+"""
+Train post-hoc SVMs using the algorithm and hyper-parameters from
+traditional R-CNN.
+"""
+
+import _init_paths
+from fast_rcnn.config import cfg, cfg_from_file
+from datasets.factory import get_imdb
+from fast_rcnn.test import im_detect
+from utils.timer import Timer
+import caffe
+import argparse
+import pprint
+import numpy as np
+import numpy.random as npr
+import cv2
+from sklearn import svm
+import os, sys
+
+class SVMTrainer(object):
+    """
+    Trains post-hoc detection SVMs for all classes using the algorithm
+    and hyper-parameters of traditional R-CNN.
+    """
+
+    def __init__(self, net, imdb):
+        self.imdb = imdb
+        self.net = net
+        self.layer = 'fc7'
+        self.hard_thresh = -1.0001
+        self.neg_iou_thresh = 0.3
+
+        dim = net.params['cls_score'][0].data.shape[1]
+        scale = self._get_feature_scale()
+        print('Feature dim: {}'.format(dim))
+        print('Feature scale: {:.3f}'.format(scale))
+        self.trainers = [SVMClassTrainer(cls, dim, feature_scale=scale)
+                         for cls in imdb.classes]
+
+    def _get_feature_scale(self, num_images=100):
+        TARGET_NORM = 20.0 # Magic value from traditional R-CNN
+        _t = Timer()
+        roidb = self.imdb.roidb
+        total_norm = 0.0
+        count = 0.0
+        inds = npr.choice(xrange(self.imdb.num_images), size=num_images,
+                          replace=False)
+        for i_, i in enumerate(inds):
+            im = cv2.imread(self.imdb.image_path_at(i))
+            if roidb[i]['flipped']:
+                im = im[:, ::-1, :]
+            _t.tic()
+            scores, boxes = im_detect(self.net, im, roidb[i]['boxes'])
+            _t.toc()
+            feat = self.net.blobs[self.layer].data
+            total_norm += np.sqrt((feat ** 2).sum(axis=1)).sum()
+            count += feat.shape[0]
+            print('{}/{}: avg feature norm: {:.3f}'.format(i_ + 1, num_images,
+                                                           total_norm / count))
+
+        return TARGET_NORM * 1.0 / (total_norm / count)
+
+    def _get_pos_counts(self):
+        counts = np.zeros((len(self.imdb.classes)), dtype=np.int)
+        roidb = self.imdb.roidb
+        for i in xrange(len(roidb)):
+            for j in xrange(1, self.imdb.num_classes):
+                I = np.where(roidb[i]['gt_classes'] == j)[0]
+                counts[j] += len(I)
+
+        for j in xrange(1, self.imdb.num_classes):
+            print('class {:s} has {:d} positives'.
+                  format(self.imdb.classes[j], counts[j]))
+
+        return counts
+
+    def get_pos_examples(self):
+        counts = self._get_pos_counts()
+        for i in xrange(len(counts)):
+            self.trainers[i].alloc_pos(counts[i])
+
+        _t = Timer()
+        roidb = self.imdb.roidb
+        num_images = len(roidb)
+        # num_images = 100
+        for i in xrange(num_images):
+            im = cv2.imread(self.imdb.image_path_at(i))
+            if roidb[i]['flipped']:
+                im = im[:, ::-1, :]
+            gt_inds = np.where(roidb[i]['gt_classes'] > 0)[0]
+            gt_boxes = roidb[i]['boxes'][gt_inds]
+            _t.tic()
+            scores, boxes = im_detect(self.net, im, gt_boxes)
+            _t.toc()
+            feat = self.net.blobs[self.layer].data
+            for j in xrange(1, self.imdb.num_classes):
+                cls_inds = np.where(roidb[i]['gt_classes'][gt_inds] == j)[0]
+                if len(cls_inds) > 0:
+                    cls_feat = feat[cls_inds, :]
+                    self.trainers[j].append_pos(cls_feat)
+
+            print 'get_pos_examples: {:d}/{:d} {:.3f}s' \
+                  .format(i + 1, len(roidb), _t.average_time)
+
+    def initialize_net(self):
+        # Start all SVM parameters at zero
+        self.net.params['cls_score'][0].data[...] = 0
+        self.net.params['cls_score'][1].data[...] = 0
+
+        # Initialize SVMs in a smart way. Not doing this because its such
+        # a good initialization that we might not learn something close to
+        # the SVM solution.
+#        # subtract background weights and biases for the foreground classes
+#        w_bg = self.net.params['cls_score'][0].data[0, :]
+#        b_bg = self.net.params['cls_score'][1].data[0]
+#        self.net.params['cls_score'][0].data[1:, :] -= w_bg
+#        self.net.params['cls_score'][1].data[1:] -= b_bg
+#        # set the background weights and biases to 0 (where they shall remain)
+#        self.net.params['cls_score'][0].data[0, :] = 0
+#        self.net.params['cls_score'][1].data[0] = 0
+
+    def update_net(self, cls_ind, w, b):
+        self.net.params['cls_score'][0].data[cls_ind, :] = w
+        self.net.params['cls_score'][1].data[cls_ind] = b
+
+    def train_with_hard_negatives(self):
+        _t = Timer()
+        roidb = self.imdb.roidb
+        num_images = len(roidb)
+        # num_images = 100
+        for i in xrange(num_images):
+            im = cv2.imread(self.imdb.image_path_at(i))
+            if roidb[i]['flipped']:
+                im = im[:, ::-1, :]
+            _t.tic()
+            scores, boxes = im_detect(self.net, im, roidb[i]['boxes'])
+            _t.toc()
+            feat = self.net.blobs[self.layer].data
+            for j in xrange(1, self.imdb.num_classes):
+                hard_inds = \
+                    np.where((scores[:, j] > self.hard_thresh) &
+                             (roidb[i]['gt_overlaps'][:, j].toarray().ravel() <
+                              self.neg_iou_thresh))[0]
+                if len(hard_inds) > 0:
+                    hard_feat = feat[hard_inds, :].copy()
+                    new_w_b = \
+                        self.trainers[j].append_neg_and_retrain(feat=hard_feat)
+                    if new_w_b is not None:
+                        self.update_net(j, new_w_b[0], new_w_b[1])
+
+            print(('train_with_hard_negatives: '
+                   '{:d}/{:d} {:.3f}s').format(i + 1, len(roidb),
+                                               _t.average_time))
+
+    def train(self):
+        # Initialize SVMs using
+        #   a. w_i = fc8_w_i - fc8_w_0
+        #   b. b_i = fc8_b_i - fc8_b_0
+        #   c. Install SVMs into net
+        self.initialize_net()
+
+        # Pass over roidb to count num positives for each class
+        #   a. Pre-allocate arrays for positive feature vectors
+        # Pass over roidb, computing features for positives only
+        self.get_pos_examples()
+
+        # Pass over roidb
+        #   a. Compute cls_score with forward pass
+        #   b. For each class
+        #       i. Select hard negatives
+        #       ii. Add them to cache
+        #   c. For each class
+        #       i. If SVM retrain criteria met, update SVM
+        #       ii. Install new SVM into net
+        self.train_with_hard_negatives()
+
+        # One final SVM retraining for each class
+        # Install SVMs into net
+        for j in xrange(1, self.imdb.num_classes):
+            new_w_b = self.trainers[j].append_neg_and_retrain(force=True)
+            self.update_net(j, new_w_b[0], new_w_b[1])
+
+class SVMClassTrainer(object):
+    """Manages post-hoc SVM training for a single object class."""
+
+    def __init__(self, cls, dim, feature_scale=1.0,
+                 C=0.001, B=10.0, pos_weight=2.0):
+        self.pos = np.zeros((0, dim), dtype=np.float32)
+        self.neg = np.zeros((0, dim), dtype=np.float32)
+        self.B = B
+        self.C = C
+        self.cls = cls
+        self.pos_weight = pos_weight
+        self.dim = dim
+        self.feature_scale = feature_scale
+        self.svm = svm.LinearSVC(C=C, class_weight={1: 2, -1: 1},
+                                 intercept_scaling=B, verbose=1,
+                                 penalty='l2', loss='l1',
+                                 random_state=cfg.RNG_SEED, dual=True)
+        self.pos_cur = 0
+        self.num_neg_added = 0
+        self.retrain_limit = 2000
+        self.evict_thresh = -1.1
+        self.loss_history = []
+
+    def alloc_pos(self, count):
+        self.pos_cur = 0
+        self.pos = np.zeros((count, self.dim), dtype=np.float32)
+
+    def append_pos(self, feat):
+        num = feat.shape[0]
+        self.pos[self.pos_cur:self.pos_cur + num, :] = feat
+        self.pos_cur += num
+
+    def train(self):
+        print('>>> Updating {} detector <<<'.format(self.cls))
+        num_pos = self.pos.shape[0]
+        num_neg = self.neg.shape[0]
+        print('Cache holds {} pos examples and {} neg examples'.
+              format(num_pos, num_neg))
+        X = np.vstack((self.pos, self.neg)) * self.feature_scale
+        y = np.hstack((np.ones(num_pos),
+                       -np.ones(num_neg)))
+        self.svm.fit(X, y)
+        w = self.svm.coef_
+        b = self.svm.intercept_[0]
+        scores = self.svm.decision_function(X)
+        pos_scores = scores[:num_pos]
+        neg_scores = scores[num_pos:]
+
+        pos_loss = (self.C * self.pos_weight *
+                    np.maximum(0, 1 - pos_scores).sum())
+        neg_loss = self.C * np.maximum(0, 1 + neg_scores).sum()
+        reg_loss = 0.5 * np.dot(w.ravel(), w.ravel()) + 0.5 * b ** 2
+        tot_loss = pos_loss + neg_loss + reg_loss
+        self.loss_history.append((tot_loss, pos_loss, neg_loss, reg_loss))
+
+        for i, losses in enumerate(self.loss_history):
+            print(('    {:d}: obj val: {:.3f} = {:.3f} '
+                   '(pos) + {:.3f} (neg) + {:.3f} (reg)').format(i, *losses))
+
+        # Sanity check
+        scores_ret = (
+                X * 1.0 / self.feature_scale).dot(w.T * self.feature_scale) + b
+        assert np.allclose(scores, scores_ret[:, 0], atol=1e-5), \
+                "Scores from returned model don't match decision function"
+
+        return ((w * self.feature_scale, b), pos_scores, neg_scores)
+
+    def append_neg_and_retrain(self, feat=None, force=False):
+        if feat is not None:
+            num = feat.shape[0]
+            self.neg = np.vstack((self.neg, feat))
+            self.num_neg_added += num
+        if self.num_neg_added > self.retrain_limit or force:
+            self.num_neg_added = 0
+            new_w_b, pos_scores, neg_scores = self.train()
+            # scores = np.dot(self.neg, new_w_b[0].T) + new_w_b[1]
+            # easy_inds = np.where(neg_scores < self.evict_thresh)[0]
+            not_easy_inds = np.where(neg_scores >= self.evict_thresh)[0]
+            if len(not_easy_inds) > 0:
+                self.neg = self.neg[not_easy_inds, :]
+                # self.neg = np.delete(self.neg, easy_inds)
+            print('    Pruning easy negatives')
+            print('    Cache holds {} pos examples and {} neg examples'.
+                  format(self.pos.shape[0], self.neg.shape[0]))
+            print('    {} pos support vectors'.format((pos_scores <= 1).sum()))
+            print('    {} neg support vectors'.format((neg_scores >= -1).sum()))
+            return new_w_b
+        else:
+            return None
+
+def parse_args():
+    """
+    Parse input arguments
+    """
+    parser = argparse.ArgumentParser(description='Train SVMs (old skool)')
+    parser.add_argument('--gpu', dest='gpu_id', help='GPU device id to use [0]',
+                        default=0, type=int)
+    parser.add_argument('--def', dest='prototxt',
+                        help='prototxt file defining the network',
+                        default=None, type=str)
+    parser.add_argument('--net', dest='caffemodel',
+                        help='model to test',
+                        default=None, type=str)
+    parser.add_argument('--cfg', dest='cfg_file',
+                        help='optional config file', default=None, type=str)
+    parser.add_argument('--imdb', dest='imdb_name',
+                        help='dataset to train on',
+                        default='voc_2007_trainval', type=str)
+
+    if len(sys.argv) == 1:
+        parser.print_help()
+        sys.exit(1)
+
+    args = parser.parse_args()
+    return args
+
+if __name__ == '__main__':
+    # Must turn this off to prevent issues when digging into the net blobs to
+    # pull out features (tricky!)
+    cfg.DEDUP_BOXES = 0
+
+    # Must turn this on because we use the test im_detect() method to harvest
+    # hard negatives
+    cfg.TEST.SVM = True
+
+    args = parse_args()
+
+    print('Called with args:')
+    print(args)
+
+    if args.cfg_file is not None:
+        cfg_from_file(args.cfg_file)
+
+    print('Using config:')
+    pprint.pprint(cfg)
+
+    # fix the random seed for reproducibility
+    np.random.seed(cfg.RNG_SEED)
+
+    # set up caffe
+    caffe.set_mode_gpu()
+    if args.gpu_id is not None:
+        caffe.set_device(args.gpu_id)
+    net = caffe.Net(args.prototxt, args.caffemodel, caffe.TEST)
+    net.name = os.path.splitext(os.path.basename(args.caffemodel))[0]
+    out = os.path.splitext(os.path.basename(args.caffemodel))[0] + '_svm'
+    out_dir = os.path.dirname(args.caffemodel)
+
+    imdb = get_imdb(args.imdb_name)
+    print 'Loaded dataset `{:s}` for training'.format(imdb.name)
+
+    # enhance roidb to contain flipped examples
+    if cfg.TRAIN.USE_FLIPPED:
+        print 'Appending horizontally-flipped training examples...'
+        imdb.append_flipped_images()
+        print 'done'
+
+    SVMTrainer(net, imdb).train()
+
+    filename = '{}/{}.caffemodel'.format(out_dir, out)
+    net.save(filename)
+    print 'Wrote svm model to: {:s}'.format(filename)