-
Notifications
You must be signed in to change notification settings - Fork 29
/
training_run_notes_spot.txt
3789 lines (3206 loc) · 372 KB
/
training_run_notes_spot.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
Tab 7: ~/src/CoppeliaSim_Edu_V4_0_0_Ubuntu18_04/coppeliaSim.sh -gREMOTEAPISERVERSERVICE_19990_FALSE_TRUE -s ~/src/real_good_robot/simulation/simulation.ttt
Tab 8: ~/src/CoppeliaSim_Edu_V4_0_0_Ubuntu18_04/coppeliaSim.sh -gREMOTEAPISERVERSERVICE_19998_FALSE_TRUE -s ~/src/real_good_robot/simulation/simulation.ttt
Tab 9: ~/src/CoppeliaSim_Edu_V4_0_0_Ubuntu18_04/coppeliaSim.sh -gREMOTEAPISERVERSERVICE_19999_FALSE_TRUE -s ~/src/real_good_robot/simulation/simulation.ttt
Tab 10: ~/src/CoppeliaSim_Edu_V4_0_0_Ubuntu18_04/coppeliaSim.sh -gREMOTEAPISERVERSERVICE_20000_FALSE_TRUE -s ~/src/real_good_robot/simulation/simulation.ttt
SIM STACK - SPOT STANDARD Trial rtrial Task Progress - TRIAL REWARD - RANDOM ACTIONS - SORT TRIAL REWARD - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-05-28
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --trial_reward --check_z_height --tcp_port 19990 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions
RESUME: ± export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --trial_reward --check_z_height --tcp_port 19990 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --resume /home/ahundt/src/real_good_robot/logs/2020-05-28-12-09-32_Sim-Stack-SPOT-Trial-Reward-Training
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-05-28-12-09-32_Sim-Stack-SPOT-Trial-Reward-Training
Commit: a534735959ec2747c3b134a6d3067135a5c7bd75 release tag:v0.16.0
GPU 0, Tab 0, port 19990, left v-rep window, v-rep tab 7
CANCELLED BECAUSE IT IS NOT THE RUN WE NEED RIGHT NOW - 2020-05-30 - TODO MAYBE RESUME LATER, finished around 7k actions
SIM ROW - SPOT STANDARD Trial rtrial Task Progress - TRIAL REWARD - RANDOM ACTIONS - SORT TRIAL REWARD - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-05-28
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="1" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --trial_reward --check_row --tcp_port 19998 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions
RESUME: export CUDA_VISIBLE_DEVICES="1" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --trial_reward --check_row --tcp_port 19998 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --resume /home/ahundt/src/real_good_robot/logs/2020-05-28-12-10-18_Sim-Rows-SPOT-Trial-Reward-Training
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-05-28-12-10-18_Sim-Rows-SPOT-Trial-Reward-Training
Commit: a534735959ec2747c3b134a6d3067135a5c7bd75 release tag:v0.16.0
GPU 1, Tab 1, port 19998, center left v-rep window, v-rep tab 8
CANCELLED BECAUSE IT IS NOT THE RUN WE NEED RIGHT NOW - 2020-05-30 - TODO MAYBE RESUME LATER, finished around 7k actions - Resumed june 2020
RESUME: GPU 3, Tab 3, port 20000, right v-rep window, v-rep tab 10
RESUME Commit: 36a0c6a8cfd6c0d8a087f0b647814575054faedd release tag: v0.16.3
RESUME: export CUDA_VISIBLE_DEVICES="3" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --trial_reward --check_row --tcp_port 20000 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --resume /home/ahundt/src/real_good_robot/logs/2020-05-28-12-10-18_Sim-Rows-SPOT-Trial-Reward-Training
> '/home/ahundt/src/real_good_robot/logs/2020-05-28-12-10-18_Sim-Rows-SPOT-Trial-Reward-Training/2020-06-11-00-40-54_Sim-Rows-SPOT-Trial-Reward-Testing/best_stats.json'
> {"action_efficiency_best_index": 1779, "action_efficiency_best_value": 0.29375351716375914, "grasp_success_rate_best_index": 1778, "grasp_success_rate_best_value": 0.33109619686800895, "place_success_rate_best_index": null, "place_success_rate_best_value": -Infinity, "trial_success_rate_best_index": 1777, "trial_success_rate_best_value": 0.74}
>
> place success rate log value: 8.744292237442922167e-01
> Manually edited json with place success rate :
> '/home/ahundt/src/real_good_robot/logs/2020-05-28-12-10-18_Sim-Rows-SPOT-Trial-Reward-Training/2020-06-11-00-40-54_Sim-Rows-SPOT-Trial-Reward-Testing/best_stats.json'
> {"action_efficiency_best_index": 1779, "action_efficiency_best_value": 0.29375351716375914, "grasp_success_rate_best_index": 1778, "grasp_success_rate_best_value": 0.33109619686800895, "place_success_rate_best_index": null, "place_success_rate_best_value": 0.874, "trial_success_rate_best_index": 1777, "trial_success_rate_best_value": 0.74}
SIM STACK - Trial Situation Removal - RANDOM ACTIONS - SORT TRIAL REWARD - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-05-28
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="2" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --trial_reward --check_z_height --tcp_port 19999 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --no_height_reward
RESUME: export CUDA_VISIBLE_DEVICES="2" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --trial_reward --check_z_height --tcp_port 19999 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --no_height_reward --resume /home/ahundt/src/real_good_robot/logs/2020-05-28-12-11-00_Sim-Stack-SPOT-Trial-Reward-Training
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-05-28-12-11-00_Sim-Stack-SPOT-Trial-Reward-Training
Commit: a534735959ec2747c3b134a6d3067135a5c7bd75 release tag:v0.16.0
GPU 2, Tab 2, port 19999, left v-rep window, v-rep tab 9
CANCELLED BECAUSE IT IS NOT THE RUN WE NEED RIGHT NOW - 2020-05-30 - TODO MAYBE RESUME LATER, finished around 7k actions
SIM ROW - Trial Situation Removal - RANDOM ACTIONS - SORT TRIAL REWARD - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-05-28
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="3" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --trial_reward --check_row --tcp_port 20000 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --no_height_reward
RESUME: ± export CUDA_VISIBLE_DEVICES="3" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --trial_reward --check_row --tcp_port 20000 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --no_height_reward --resume /home/ahundt/src/real_good_robot/logs/2020-05-28-12-11-38_Sim-Rows-SPOT-Trial-Reward-Training
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-05-28-12-11-38_Sim-Rows-SPOT-Trial-Reward-Training
Commit: a534735959ec2747c3b134a6d3067135a5c7bd75 release tag:v0.16.0
GPU 3, Tab 3, port 19999, center left v-rep window, v-rep tab 10
CANCELLED BECAUSE IT IS NOT THE RUN WE NEED RIGHT NOW - 2020-05-30 - TODO MAYBE RESUME LATER, finished around 7k actions
SIM ROW - Trial Situation Removal - RANDOM ACTIONS - SORT TRIAL REWARD - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-05-30
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="3" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --check_row --tcp_port 20000 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --no_height_reward
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-05-30-13-12-47_Sim-Rows-SPOT-Trial-Reward-Training
Commit: 12d9481717486342dbfcaff191ddb1428f102406 release tag:v0.16.1
GPU 3, Tab 3, port 19999, center left v-rep window, v-rep tab 10
CANCELLED BECAUSE IT IS NOT THE RUN WE NEED RIGHT NOW - 2020-05-30 - TODO MAYBE RESUME LATER, finished around 1k actions
Pass 1 - Ablation of instant reward shcedules
============================================================================================================================
SIM STACK - Task Progress aka progress only - RANDOM ACTIONS - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-05-30
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --check_z_height --tcp_port 19990 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-05-30-13-09-38_Sim-Stack-Two-Step-Reward-Training
Commit: 12d9481717486342dbfcaff191ddb1428f102406 release tag:v0.16.1
GPU 0, Tab 0, port 19990, left v-rep window, v-rep tab 7
> Trial logging complete: 101 --------------------------------------------------------------
> Running two step backprop()
> Primitive confidence scores: 3.220978 (push), 7.850247 (grasp), 7.190622 (place)
> Action: grasp at (4, 7, 152)
> Training loss: 0.672980
> Executing: grasp at (-0.420000, -0.210000, 0.001003) orientation: 1.570796
> gripper position: 0.0304451584815979
> gripper position: 0.026506200432777405
> gripper position: 0.0013817846775054932
> gripper position: -0.022582605481147766
> gripper position: -0.04284219443798065
> Grasp successful: False
> prev_height: 0.0 max_z: 0.05112840983826451 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> prev_height: 1.0 max_z: 1.0225681967652902 goal_success: False needed to reset: False max_workspace_height: 0.6 <<<<<<<<<<<
> check_stack() stack_height: 1.0225681967652902 stack matches current goal: False partial_stack_success: False Does the code think a reset
> is needed: False
> STACK: trial: 101 actions/partial: 3.662721893491124 actions/full stack: 12.505050505050505 (lower is better) Grasp Count: 663, grasp success rate: 0.8310708898944194 place_on_stack_rate: 0.6134301270417423 place_attempts: 551 partial_stack_successes: 338 stack_successes: 99 trial_success_rate: 0.9801980198019802 stack goal: None current_height: 1.0225681967652902
> trial_complete_indices: [ 13. 19. 32. 44. 68. 76. 86. 102. 112. 118. 125. 133.
> 139. 145. 180. 186. 192. 198. 205. 213. 221. 232. 240. 293.
> 299. 309. 317. 323. 333. 349. 354. 365. 376. 380. 393. 403.
> 420. 456. 476. 492. 496. 504. 515. 521. 531. 537. 543. 551.
> 560. 575. 585. 591. 600. 610. 620. 631. 639. 645. 652. 658.
> 672. 704. 711. 717. 721. 729. 746. 758. 764. 770. 778. 784.
> 790. 796. 806. 812. 855. 861. 897. 905. 915. 919. 928. 962.
> 968. 972. 979. 987. 993. 997. 1007. 1013. 1031. 1043. 1149. 1160.
> 1168. 1178. 1195. 1231. 1237.]
> Max trial success rate: 0.98, at action iteration: 1234. (total of 1236 actions, max excludes first 1234 actions)
> Max grasp success rate: 0.8335854765506808, at action iteration: 1235. (total of 1236 actions, max excludes first 1234 actions)
> Max place success rate: 0.7426086956521739, at action iteration: 1236. (total of 1237 actions, max excludes first 1234 actions)
> Max action efficiency: 0.5153970826580226, at action iteration: 1236. (total of 1237 actions, max excludes first 1234 actions)
> saving plot: 2020-06-03-03-08-00_Sim-Stack-Two-Step-Reward-Testing-Sim-Stack-Two-Step-Reward-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-03-03-08-00_Sim-Stack-Two-Step-Reward-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-03-03-08-00_Sim-Stack-Two-Step-Reward-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-05-30-13-09-38_Sim-Stack-Two-Step-Reward-Training/2020-06-03-03-08-00_Sim-Stack-Two-Step-Reward-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 0.98, 'trial_success_rate_best_index': 1234, 'grasp_success_rate_best_value': 0.8335854765506808, 'grasp_success_rate_best_index': 1235, 'place_success_rate_best_value': 0.7426086956521739, 'place_success_rate_best_index': 1236, 'action_efficiency_best_value': 0.5153970826580226, 'action_efficiency_best_index': 1236}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-05-30-13-09-38_Sim-Stack-Two-Step-Reward-Training
> Training results:
> {'trial_success_rate_best_value': 0.8103448275862069, 'trial_success_rate_best_index': 19569, 'grasp_success_rate_best_value': 0.9494163424124513, 'grasp_success_rate_best_index': 12019, 'place_success_rate_best_value': 0.8312236286919831, 'place_success_rate_best_index': 17156, 'action_efficiency_best_value': 0.6, 'action_efficiency_best_index': 19471}
SIM ROW - Task Progress aka progress only - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-05-30
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="1" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --check_row --tcp_port 19998 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-05-30-13-10-52_Sim-Rows-Two-Step-Reward-Training
Commit: 12d9481717486342dbfcaff191ddb1428f102406 release tag:v0.16.1
GPU 1, Tab 1, port 19998, center left v-rep window, v-rep tab 8
> Trial logging complete: 101 --------------------------------------------------------------
> Running two step backprop()
> Primitive confidence scores: 5.558454 (push), 8.079895 (grasp), 9.956761 (place)
> Action: grasp at (4, 183, 167)
> Training loss: 0.173360
> Executing: grasp at (-0.390000, 0.142000, 0.001004) orientation: 1.570796
> gripper position: 0.030987784266471863
> gripper position: 0.02650594152510166
> gripper position: 0.0014807581901550293
> gripper position: -0.023117437958717346
> gripper position: -0.042321473360061646
> Grasp successful: False
> prev_height: 0.0 max_z: 0.05110803083189876 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> check_row: True | row_size: 2 | blocks: ['blue' 'red']
> check_stack() stack_height: 2 stack matches current goal: True partial_stack_success: True Does the code think a reset is needed: False
> STACK: trial: 101 actions/partial: 6.386363636363637 actions/full stack: 11.46938775510204 (lower is better) Grasp Count: 618, grasp success rate: 0.8220064724919094 place_on_stack_rate: 0.34782608695652173 place_attempts: 506 partial_stack_successes: 176 stack_successes: 98 trial_success_rate: 0.9702970297029703 stack goal: [3 2] current_height: 2
> trial_complete_indices: [ 4. 8. 12. 18. 23. 28. 32. 36. 40. 44. 50. 54.
> 58. 62. 66. 68. 72. 76. 80. 84. 485. 491. 499. 503.
> 509. 511. 517. 521. 525. 708. 712. 714. 720. 728. 730. 736.
> 740. 742. 749. 753. 758. 770. 776. 778. 786. 790. 794. 801.
> 805. 809. 811. 813. 815. 819. 828. 832. 836. 840. 845. 849.
> 861. 865. 869. 872. 878. 882. 886. 888. 890. 894. 896. 900.
> 904. 950. 955. 957. 963. 968. 976. 980. 982. 988. 991. 993.
> 999. 1044. 1053. 1057. 1064. 1068. 1074. 1078. 1092. 1096. 1100. 1104.
> 1106. 1110. 1114. 1118. 1123.]
> Max trial success rate: 0.97, at action iteration: 1120. (total of 1122 actions, max excludes first 1120 actions)
> Max grasp success rate: 0.823051948051948, at action iteration: 1120. (total of 1122 actions, max excludes first 1120 actions)
> Max place success rate: 0.8950495049504951, at action iteration: 1120. (total of 1123 actions, max excludes first 1120 actions)
> Max action efficiency: 0.5303571428571429, at action iteration: 1122. (total of 1123 actions, max excludes first 1120 actions)
> saving plot: 2020-06-03-07-00-00_Sim-Rows-Two-Step-Reward-Testing-Sim-Rows-Two-Step-Reward-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-03-07-00-00_Sim-Rows-Two-Step-Reward-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-03-07-00-00_Sim-Rows-Two-Step-Reward-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-05-30-13-10-52_Sim-Rows-Two-Step-Reward-Training/2020-06-03-07-00-00_Sim-Rows-Two-Step-Reward-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 0.97, 'trial_success_rate_best_index': 1120, 'grasp_success_rate_best_value': 0.823051948051948, 'grasp_success_rate_best_index': 1120, 'place_success_rate_best_value': 0.8950495049504951, 'place_success_rate_best_index': 1120, 'action_efficiency_best_value': 0.5303571428571429, 'action_efficiency_best_index': 1122}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-05-30-13-10-52_Sim-Rows-Two-Step-Reward-Training
> Training results:
> {'action_efficiency_best_index': 18908, 'action_efficiency_best_value': 1.248, 'grasp_success_rate_best_index': 17217, 'grasp_success_rate_best_value': 0.8145454545454546, 'place_success_rate_best_index': 13436, 'place_success_rate_best_value': 0.9345794392523364, 'trial_success_rate_best_index': 18572, 'trial_success_rate_best_value': 0.768}
SIM STACK - Basic Situation Removal - RANDOM ACTIONS - SORT TRIAL REWARD - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-05-30
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="2" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --check_z_height --tcp_port 19999 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --no_height_reward
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-05-30-13-11-57_Sim-Stack-Two-Step-Reward-Training
Commit: 12d9481717486342dbfcaff191ddb1428f102406 release tag:v0.16.1
GPU 2, Tab 2, port 19999, left v-rep window, v-rep tab 9
> '/home/ahundt/src/real_good_robot/logs/2020-05-30-13-11-57_Sim-Stack-Two-Step-Reward-Training/2020-06-02-19-24-53_Sim-Stack-Two-Step-Reward-Testing'
> {"action_efficiency_best_index": 7161, "action_efficiency_best_value": 0.07710574102528286, "grasp_success_rate_best_index": 7159, "grasp_success_rate_best_value": 0.8780807551127425, "place_success_rate_best_index": 7159, "place_success_rate_best_value": 0.6434548714883442, "trial_success_rate_best_index": 7159, "trial_success_rate_best_value": 0.9}
SIM ROW - Basic Situation Removal - RANDOM ACTIONS - SORT TRIAL REWARD - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-05-30
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="3" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --check_row --tcp_port 20000 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --no_height_reward
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-05-30-17-46-01_Sim-Rows-Two-Step-Reward-Training
Commit: 12d9481717486342dbfcaff191ddb1428f102406 release tag:v0.16.1
GPU 3, Tab 3, port 20000, center left v-rep window, v-rep tab 10
> Complete first test run trial success rate best value model:
> {"action_efficiency_best_index": 2092, "action_efficiency_best_value": 0.28995215311004785, "grasp_success_rate_best_index": 2090, "grasp_success_rate_best_value": 0.8680926916221033, "place_success_rate_best_index": 2090, "place_success_rate_best_value": 0.5927835051546392, "trial_success_rate_best_index": 2090, "trial_success_rate_best_value": 0.94}
> Max trial success rate: 0.94, at action iteration: 2090. (total of 2092 actions, max excludes first 2090 actions)
> Max grasp success rate: 0.8680926916221033, at action iteration: 2090. (total of 2092 actions, max excludes first 2090 actions) Max place success rate: 0.5927835051546392, at action iteration: 2090. (total of 2091 actions, max excludes first 2090 actions)
> Max action efficiency: 0.28995215311004785, at action iteration: 2092. (total of 2093 actions, max excludes first 2090 actions)
> saving plot: 2020-06-03-06-28-31_Sim-Rows-Two-Step-Reward-Testing-Sim-Rows-Two-Step-Reward-Testing_success_plot.png saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-03-06-28-31_Sim-Rows-Two-Step-Reward-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-03-06-28-31_Sim-Rows-Two-Step-Reward-Testing/best_stats.json
> Trial logging complete: 101 --------------------------------------------------------------
> *Partially* complete second test run, best action efficiency model:
> -------------------------------------
> Max trial success rate: 0.9710144927536232, at action iteration: 1741. (total of 1743 actions, max excludes first 1741 actions)
> Max grasp success rate: 0.9186176142697882, at action iteration: 1741. (total of 1743 actions, max excludes first 1741 actions)
> Max place success rate: 0.6415552855407047, at action iteration: 1741. (total of 1742 actions, max excludes first 1741 actions)
> Max action efficiency: 0.24813325674899483, at action iteration: 1743. (total of 1744 actions, max excludes first 1741 actions)
> saving plot: 2020-06-03-15-46-00_Sim-Rows-Two-Step-Reward-Testing-Sim-Rows-Two-Step-Reward-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-03-15-46-00_Sim-Rows-Two-Step-Reward-Testing/data/best_stats.json saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-03-15-46-00_Sim-Rows-Two-Step-Reward-Testing/best_stats.json
> Trial logging complete: 70 --------------------------------------------------------------
Pass 2 - SPOT-Q TASK PROGRESS MASKED
===========================================================================================================================
SIM STACK - Task Progress SPOT-Q MASKED - RANDOM ACTIONS - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-06-03
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --check_z_height --tcp_port 19990 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-03-11-44-02_Sim-Stack-Two-Step-Reward-Masked-Training
Commit: 12d9481717486342dbfcaff191ddb1428f102406 release tag:v0.16.1
GPU 0, Tab 0, port 19990, left v-rep window, v-rep tab 7
> Trial logging complete: 101 --------------------------------------------------------------
> prev_height: 0.0 max_z: 0.05115434739934034 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> prev_height: 1.0 max_z: 1.0230869479868068 goal_success: False needed to reset: False max_workspace_height: 0.6 <<<<<<<<<<<
> check_stack() stack_height: 1.0230869479868068 stack matches current goal: False partial_stack_success: False Does the code think a reset
> is needed: False
> STACK: trial: 101 actions/partial: 3.8494318181818183 actions/full stack: 13.415841584158416 (lower is better) Grasp Count: 768, grasp
> success rate: 0.7552083333333334 place_on_stack_rate: 0.6068965517241379 place_attempts: 580 partial_stack_successes: 352 stack_successes: 101 trial_success_rate: 1.0 stack goal: None current_height: 1.0230869479868068
> trial_complete_indices: [ 7. 20. 30. 36. 45. 74. 84. 99. 118. 138. 167. 177.
> 189. 213. 225. 249. 255. 269. 275. 283. 290. 305. 323. 327.
> 341. 351. 360. 391. 403. 420. 446. 457. 472. 481. 505. 534.
> 543. 553. 562. 573. 592. 600. 606. 620. 626. 658. 664. 674.
> 680. 689. 697. 705. 709. 726. 737. 745. 760. 764. 770. 780.
> 791. 827. 850. 862. 878. 897. 905. 917. 930. 958. 971. 982.
> 1022. 1028. 1034. 1047. 1062. 1069. 1078. 1084. 1102. 1109. 1114. 1133.
> 1145. 1165. 1173. 1184. 1221. 1233. 1241. 1252. 1275. 1282. 1290. 1296.
> 1312. 1319. 1333. 1342. 1354.]
> Max trial success rate: 1.0, at action iteration: 1351. (total of 1353 actions, max excludes first 1351 actions)
> Max grasp success rate: 0.7558746736292428, at action iteration: 1351. (total of 1353 actions, max excludes first 1351 actions)
> Max place success rate: 0.757679180887372, at action iteration: 1351. (total of 1354 actions, max excludes first 1351 actions)
> Max action efficiency: 0.44855662472242785, at action iteration: 1353. (total of 1354 actions, max excludes first 1351 actions)
> saving plot: 2020-06-07-06-26-25_Sim-Stack-Two-Step-Reward-Masked-Testing-Sim-Stack-Two-Step-Reward-Masked-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-07-06-26-25_Sim-Stack-Two-Step-Reward-Masked-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-07-06-26-25_Sim-Stack-Two-Step-Reward-Masked-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-03-11-44-02_Sim-Stack-Two-Step-Reward-Masked-Training/2020-06-07-06-26-25_Sim-Stack-Two-Step-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 1351, 'grasp_success_rate_best_value': 0.7558746736292428, 'grasp_success_rate_best_index': 1351, 'place_success_rate_best_value': 0.757679180887372, 'place_success_rate_best_index': 1351, 'action_efficiency_best_value': 0.44855662472242785, 'action_efficiency_best_index': 1353}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-03-11-44-02_Sim-Stack-Two-Step-Reward-Masked-Training
> Training results:
> {'trial_success_rate_best_value': 0.7931034482758621, 'trial_success_rate_best_index': 12797, 'grasp_success_rate_best_value': 0.937984496124031, 'grasp_success_rate_best_index': 13126, 'place_success_rate_best_value': 0.8201754385964912, 'place_success_rate_best_index': 19959, 'action_efficiency_best_value': 0.576, 'action_efficiency_best_index': 12886}
SIM ROW - Task Progress SPOT-Q MASKED - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-06-03
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="1" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --check_row --tcp_port 19998 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-03-12-05-28_Sim-Rows-Two-Step-Reward-Masked-Training
Commit: 12d9481717486342dbfcaff191ddb1428f102406 release tag:v0.16.1
GPU 1, Tab 1, port 19998, center left v-rep window, v-rep tab 8
> Trial logging complete: 101 --------------------------------------------------------------
> prev_height: 0.0 max_z: 0.05113576211473993 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> check_row: True | row_size: 2 | blocks: ['blue' 'green']
> check_stack() stack_height: 2 stack matches current goal: False partial_stack_success: False Does the code think a reset is needed: True
> main.py check_stack() DETECTED PROGRESS REVERSAL, mismatch between the goal height: 3 and current workspace stack height: 2
> STACK: trial: 101 actions/partial: 3.1473214285714284 actions/full stack: 7.05 (lower is better) Grasp Count: 372, grasp success rate:
> 0.8978494623655914 place_on_stack_rate: 0.6726726726726727 place_attempts: 333 partial_stack_successes: 224 stack_successes: 100 trial_success_rate: 0.9900990099009901 stack goal: [3 2 0 1] current_height: 2
> trial_complete_indices: [ 2. 8. 12. 23. 27. 31. 35. 39. 50. 58. 60. 66. 75. 86.
> 88. 94. 103. 109. 112. 116. 122. 130. 136. 140. 146. 152. 172. 176.
> 183. 189. 193. 200. 203. 212. 219. 223. 227. 233. 239. 251. 260. 269.
> 273. 279. 287. 293. 297. 307. 313. 319. 328. 332. 338. 342. 346. 350.
> 352. 354. 358. 360. 362. 367. 375. 381. 408. 412. 415. 419. 423. 435.
> 441. 445. 449. 451. 453. 457. 461. 467. 475. 482. 491. 495. 503. 511.
> 515. 522. 524. 535. 541. 545. 551. 557. 561. 563. 567. 576. 580. 686.
> 693. 699. 704.]
> Max trial success rate: 0.99, at action iteration: 701. (total of 703 actions, max excludes first 701 actions)
> Max grasp success rate: 0.9, at action iteration: 701. (total of 703 actions, max excludes first 701 actions)
> Max place success rate: 0.7921686746987951, at action iteration: 701. (total of 704 actions, max excludes first 701 actions)
> Max action efficiency: 0.8473609129814551, at action iteration: 701. (total of 704 actions, max excludes first 701 actions)
> saving plot: 2020-06-07-00-35-36_Sim-Rows-Two-Step-Reward-Masked-Testing-Sim-Rows-Two-Step-Reward-Masked-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-07-00-35-36_Sim-Rows-Two-Step-Reward-Masked-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-07-00-35-36_Sim-Rows-Two-Step-Reward-Masked-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-03-12-05-28_Sim-Rows-Two-Step-Reward-Masked-Training/2020-06-06-21-34-07_Sim-Rows-Two-Step-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 667, 'grasp_success_rate_best_value': 0.850415512465374, 'grasp_success_rate_best_index': 667, 'place_success_rate_best_value': 0.7752442996742671, 'place_success_rate_best_index': 667, 'action_efficiency_best_value': 0.9265367316341829, 'action_efficiency_best_index': 667}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-03-12-05-28_Sim-Rows-Two-Step-Reward-Masked-Training
> Training results:
> {'trial_success_rate_best_value': 0.7475728155339806, 'trial_success_rate_best_index': 18139, 'grasp_success_rate_best_value': 0.8550185873605948, 'grasp_success_rate_best_index': 18207, 'place_success_rate_best_value': 0.8486238532110092, 'place_success_rate_best_index': 19937, 'action_efficiency_best_value': 1.224, 'action_efficiency_best_index': 19986}
SIM STACK - Task Progress SPOT-Q MASKED - RANDOM ACTIONS - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-06-03
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="2" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --check_z_height --tcp_port 19999 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-04-11-18-49_Sim-Stack-Two-Step-Reward-Masked-Training
Commit: 12d9481717486342dbfcaff191ddb1428f102406 release tag:v0.16.1
GPU 2, Tab 2, port 19999, right center v-rep window, v-rep tab 9
> note that we ran one extra test with place success rate, since it appears there was a glitch in the action efficiency records. This is a simulator bug which may worsen final results.
> '/home/ahundt/src/real_good_robot/logs/2020-06-04-11-18-49_Sim-Stack-Two-Step-Reward-Masked-Training/2020-06-07-19-54-35_Sim-Stack-Two-Step-Reward-Masked-Testing'
> {"action_efficiency_best_index": 2284, "action_efficiency_best_value": 0.25241016652059595, "grasp_success_rate_best_index": 2282, "grasp_success_rate_best_value": 0.7245283018867924, "place_success_rate_best_index": 2282, "place_success_rate_best_value": 0.6659707724425887, "trial_success_rate_best_index": 2282, "trial_success_rate_best_value": 0.94}
> Trial logging complete: 101 --------------------------------------------------------------
> Running two step backprop()
> Primitive confidence scores: 3.534616 (push), 7.419805 (grasp), 7.226346 (place)
> Action: grasp at (0, 168, 167)
> Training loss: 0.130998
> Executing: grasp at (-0.390000, 0.112000, 0.051003) orientation: 0.000000
> gripper position: 0.03104463219642639
> gripper position: 0.026297718286514282
> gripper position: 0.0010769963264465332
> gripper position: -0.022954285144805908
> gripper position: -0.04172489047050476
> Grasp successful: False
> prev_height: 0.0 max_z: 0.05112415348966427 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> prev_height: 1.0 max_z: 1.0224830697932854 goal_success: False needed to reset: False max_workspace_height: 0.6 <<<<<<<<<<<
> check_stack() stack_height: 1.0224830697932854 stack matches current goal: False partial_stack_success: False Does the code think a reset is needed: False
> STACK: trial: 101 actions/partial: 5.6658536585365855 actions/full stack: 23.948453608247423 (lower is better) Grasp Count: 1344, grasp success rate: 0.7313988095238095 place_on_stack_rate: 0.4187946884576098 place_attempts: 979 partial_stack_successes: 410 stack_successes: 97 trial_success_rate: 0.9603960396039604 stack goal: None current_height: 1.0224830697932854
> trial_complete_indices: [ 9. 17. 37. 64. 79. 90. 159. 220. 224. 232. 260. 271.
> 279. 290. 304. 314. 393. 425. 479. 483. 502. 508. 518. 533.
> 551. 582. 598. 604. 617. 632. 636. 648. 660. 692. 700. 719.
> 743. 767. 794. 809. 840. 850. 996. 1008. 1016. 1045. 1071. 1087.
> 1105. 1116. 1141. 1160. 1207. 1287. 1306. 1321. 1330. 1339. 1359. 1403.
> 1417. 1431. 1438. 1449. 1459. 1519. 1544. 1550. 1593. 1605. 1611. 1647.
> 1664. 1683. 1692. 1706. 1720. 1767. 1797. 1934. 1958. 2008. 2031. 2048.
> 2070. 2079. 2099. 2111. 2121. 2131. 2137. 2146. 2163. 2203. 2219. 2229.
> 2241. 2246. 2254. 2305. 2322.]
> Max trial success rate: 0.96, at action iteration: 2319. (total of 2321 actions, max excludes first 2319 actions)
> Max grasp success rate: 0.732488822652757, at action iteration: 2320. (total of 2321 actions, max excludes first 2319 actions)
> Max place success rate: 0.693564862104188, at action iteration: 2321. (total of 2322 actions, max excludes first 2319 actions)
> Max action efficiency: 0.25097024579560157, at action iteration: 2321. (total of 2322 actions, max excludes first 2319 actions)
> saving plot: 2020-06-08-05-43-46_Sim-Stack-Two-Step-Reward-Masked-Testing-Sim-Stack-Two-Step-Reward-Masked-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-08-05-43-46_Sim-Stack-Two-Step-Reward-Masked-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-08-05-43-46_Sim-Stack-Two-Step-Reward-Masked-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-04-11-18-49_Sim-Stack-Two-Step-Reward-Masked-Training/2020-06-08-05-43-46_Sim-Stack-Two-Step-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 0.96, 'trial_success_rate_best_index': 2319, 'grasp_success_rate_best_value': 0.732488822652757, 'grasp_success_rate_best_index': 2320, 'place_success_rate_best_value': 0.693564862104188, 'place_success_rate_best_index': 2321, 'action_efficiency_best_value': 0.25097024579560157, 'action_efficiency_best_index': 2321}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-04-11-18-49_Sim-Stack-Two-Step-Reward-Masked-Training
> Training results:
> {'trial_success_rate_best_value': 0.8125, 'trial_success_rate_best_index': 10322, 'grasp_success_rate_best_value': 0.8905660377358491, 'grasp_success_rate_best_index': 10252, 'place_success_rate_best_value': 0.8028169014084507, 'place_success_rate_best_index': 4893, 'action_efficiency_best_value': 0.792, 'action_efficiency_best_index': 12478}
SIM ROW - Task Progress SPOT-Q MASKED - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-06-03
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="3" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --check_row --tcp_port 20000 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-03-23-18-31_Sim-Rows-Two-Step-Reward-Masked-Training
Commit: 12d9481717486342dbfcaff191ddb1428f102406 release tag:v0.16.1
GPU 3, Tab 3, port 20000, right v-rep window, v-rep tab 10
> Trial logging complete: 101 --------------------------------------------------------------
> Grasp successful: False
> prev_height: 0.0 max_z: 0.051105467279345854 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> check_row: True | row_size: 3 | blocks: ['blue' 'green' 'red']
> check_stack() stack_height: 3 stack matches current goal: True partial_stack_success: True Does the code think a reset is needed: False
> STACK: trial: 101 actions/partial: 2.911214953271028 actions/full stack: 6.23 (lower is better) Grasp Count: 333, grasp success rate: 0.8738738738738738 place_on_stack_rate: 0.7379310344827587 place_attempts: 290 partial_stack_successes: 214 stack_successes: 100 trial_success_rate: 0.9900990099009901 stack goal: [0 2] current_height: 3
> trial_complete_indices: [ 4. 8. 10. 12. 17. 40. 54. 61. 65. 71. 78. 82. 87. 91.
> 95. 99. 101. 105. 111. 117. 123. 127. 131. 135. 143. 147. 151. 155.
> 159. 169. 173. 177. 181. 190. 196. 200. 206. 210. 214. 220. 225. 232.
> 240. 244. 252. 256. 260. 268. 272. 299. 303. 315. 325. 330. 334. 336.
> 343. 349. 351. 356. 364. 368. 370. 374. 378. 403. 414. 418. 420. 428.
> 432. 436. 440. 442. 446. 450. 458. 464. 479. 490. 535. 538. 542. 546.
> 554. 556. 562. 568. 577. 579. 581. 585. 589. 591. 595. 599. 605. 609.
> 616. 618. 622.]
> Max trial success rate: 0.98, at action iteration: 619. (total of 621 actions, max excludes first 619 actions)
> Max grasp success rate: 0.8761329305135952, at action iteration: 619. (total of 621 actions, max excludes first 619 actions)
> Max place success rate: 0.7612456747404844, at action iteration: 621. (total of 622 actions, max excludes first 619 actions)
> Max action efficiency: 0.9693053311793215, at action iteration: 621. (total of 622 actions, max excludes first 619 actions)
> saving plot: 2020-06-07-20-09-08_Sim-Rows-Two-Step-Reward-Masked-Testing-Sim-Rows-Two-Step-Reward-Masked-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-07-20-09-08_Sim-Rows-Two-Step-Reward-Masked-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-07-20-09-08_Sim-Rows-Two-Step-Reward-Masked-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-03-23-18-31_Sim-Rows-Two-Step-Reward-Masked-Training/2020-06-07-17-17-16_Sim-Rows-Two-Step-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 0.98, 'trial_success_rate_best_index': 614, 'grasp_success_rate_best_value': 0.9190031152647975, 'grasp_success_rate_best_index': 614, 'place_success_rate_best_value': 0.7627118644067796, 'place_success_rate_best_index': 615, 'action_efficiency_best_value': 1.01628664495114, 'action_efficiency_best_index': 616}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-03-23-18-31_Sim-Rows-Two-Step-Reward-Masked-Training
> Training results:
> {'trial_success_rate_best_value': 0.8034188034188035, 'trial_success_rate_best_index': 19062, 'grasp_success_rate_best_value': 0.8321167883211679, 'grasp_success_rate_best_index': 17961, 'place_success_rate_best_value': 0.9090909090909091, 'place_success_rate_best_index': 19959, 'action_efficiency_best_value': 1.26, 'action_efficiency_best_index': 19903}
Pass 3
============================================================
SIM ROW - Basic Situation Removal - RANDOM ACTIONS - SORT TRIAL REWARD - REWARD SCHEDULE 0.1, 1, 1 - costar 2020-06-07
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --check_row --tcp_port 19990 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --no_height_reward
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-07-14-06-58_Sim-Rows-Two-Step-Reward-Training
Commit: 84d192f5e33a8da14b5da245f6649bed9f816884 release tag:v0.16.2
GPU 0, Tab 0, port 19990, left v-rep window, v-rep tab 7
> Trial logging complete: 101 --------------------------------------------------------------
> Running two step backprop()
> Primitive confidence scores: 2.128928 (push), 2.290924 (grasp), 2.470773 (place)
> Action: grasp at (12, 56, 135)
> Training loss: 0.090713
> Executing: grasp at (-0.454000, -0.112000, 0.001002) orientation: 4.712389
> gripper position: 0.030810609459877014
> gripper position: 0.026403671130537987
> gripper position: 0.0011664032936096191
> gripper position: -0.022915594279766083
> gripper position: -0.04185757040977478
> Grasp successful: False
> prev_height: 0.0 max_z: 0.05111169094181285 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> check_row() object_color_sequence length is 0 or 1, so there is nothing to check and it passes automatically
> check_stack() stack_height: 1 stack matches current goal: True partial_stack_success: False Does the code think a reset is needed: False
> STACK: trial: 101 actions/partial: 7.6036363636363635 actions/full stack: 22.010526315789473 (lower is better) Grasp Count: 1120, grasp success rate: 0.8464285714285714 place_on_stack_rate: 0.291005291005291 place_attempts: 945 partial_stack_successes: 275 stack_successes: 95 trial_success_rate: 0.9405940594059405 stack goal: [0] current_height: 1
> trial_complete_indices: [ 6. 8. 45. 67. 303. 311. 315. 323. 348. 375. 381. 418.
> 428. 501. 503. 516. 525. 578. 584. 603. 615. 673. 697. 704.
> 710. 716. 734. 749. 755. 772. 807. 813. 825. 835. 843. 847.
> 856. 863. 867. 884. 979. 985. 996. 1002. 1008. 1014. 1033. 1037.
> 1059. 1067. 1071. 1083. 1085. 1095. 1099. 1114. 1131. 1139. 1147. 1151.
> 1157. 1163. 1175. 1193. 1199. 1214. 1258. 1264. 1289. 1302. 1310. 1312.
> 1335. 1339. 1387. 1412. 1420. 1478. 1484. 1530. 1538. 1542. 1564. 1588.
> 1625. 1642. 1650. 1654. 1656. 1704. 1741. 1828. 1841. 1872. 1896. 1954.
> 2000. 2057. 2063. 2086. 2090.]
> Max trial success rate: 0.93, at action iteration: 2087. (total of 2089 actions, max excludes first 2087 actions)
> Max grasp success rate: 0.8470483005366727, at action iteration: 2087. (total of 2089 actions, max excludes first 2087 actions)
> Max place success rate: 0.6610169491525424, at action iteration: 2087. (total of 2090 actions, max excludes first 2087 actions)
> Max action efficiency: 0.2788691902252036, at action iteration: 2089. (total of 2090 actions, max excludes first 2087 actions)
> saving trial success rate: /home/ahundt/src/real_good_robot/logs/2020-06-11-05-20-24_Sim-Rows-Two-Step-Reward-Testing/transitions/trial-success-rate.log.csv
> saving grasp success rate: /home/ahundt/src/real_good_robot/logs/2020-06-11-05-20-24_Sim-Rows-Two-Step-Reward-Testing/transitions/grasp-success-rate.log.csv
> saving place success rate: /home/ahundt/src/real_good_robot/logs/2020-06-11-05-20-24_Sim-Rows-Two-Step-Reward-Testing/transitions/place-success-rate.log.csv
> saving action efficiency: /home/ahundt/src/real_good_robot/logs/2020-06-11-05-20-24_Sim-Rows-Two-Step-Reward-Testing/transitions/action-efficiency.log.csv
> saving plot: 2020-06-11-05-20-24_Sim-Rows-Two-Step-Reward-Testing-Sim-Rows-Two-Step-Reward-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-11-05-20-24_Sim-Rows-Two-Step-Reward-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-11-05-20-24_Sim-Rows-Two-Step-Reward-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-07-14-06-58_Sim-Rows-Two-Step-Reward-Training/2020-06-10-23-16-04_Sim-Rows-Two-Step-Reward-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 0.98, 'trial_success_rate_best_index': 1456, 'grasp_success_rate_best_value': 0.8350125944584383, 'grasp_success_rate_best_index': 1457, 'place_success_rate_best_value': -inf, 'place_success_rate_best_index': None, 'action_efficiency_best_value': 0.4368131868131868, 'action_efficiency_best_index': 1458}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-07-14-06-58_Sim-Rows-Two-Step-Reward-Training
> Training results:
> {'trial_success_rate_best_value': 0.5211267605633803, 'trial_success_rate_best_index': 10892, 'grasp_success_rate_best_value': 0.7992831541218638, 'grasp_success_rate_best_index': 16239, 'place_success_rate_best_value': 0.6807511737089202, 'place_success_rate_best_index': 19740, 'action_efficiency_best_value': 0.6, 'action_efficiency_best_index': 19564}
TODO(ahundt) figure out the source of the place success rate infinite test bug. I looked it up manually in the log and the final value is 6.042296072507552518e-01 (60\%)
Parameter sensitivity experiment - SIM ROW - Task Progress SPOT-Q MASKED - REWARD SCHEDULE 1, 1, 1 - workstation named spot 2020-06-07
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="1" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --check_row --tcp_port 19998 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-07-14-07-53_Sim-Rows-Two-Step-Reward-Masked-Training
Commit: 169ee86203c2a360b14fac69bd4b5cef86de3e83 release tag:push_r_weight_1.0_v0
GPU 1, Tab 1, port 19998, center left v-rep window, v-rep tab 8
> Something really unusual happened here, as trial success wasn't recoreded correctly. Remember this is not a typical run, REWARD SCHEDULE 1, 1, 1 is changed significantly, (push 0.1 -> 1.0).
> trial_complete_indices: [ 10. 21. 26. 31. 36. 51. 62. 75. 83. 93. 112. 125.
> 140. 146. 160. 168. 179. 193. 207. 212. 222. 228. 232. 244.
> 254. 264. 274. 284. 285. 300. 309. 320. 336. 347. 350. 363.
> 372. 382. 394. 402. 407. 419. 421. 433. 443. 451. 459. 471.
> 484. 495. 506. 509. 520. 529. 535. 548. 558. 567. 576. 588.
> 591. 605. 616. 633. 645. 655. 669. 680. 703. 721. 747. 753.
> 768. 780. 792. 794. 805. 815. 823. 830. 839. 853. 869. 881.
> 891. 903. 913. 931. 946. 947. 955. 965. 973. 978. 985. 995.
> 1007. 1028. 1039. 1053.]
> /home/ahundt/src/real_good_robot/plot.py:136: RuntimeWarning: invalid value encountered in double_scalars
> var = np.sqrt(success_rate[i] * (1 - success_rate[i]) / successes.shape[0])
> Max grasp success rate: 1.0, at action iteration: 1056. (total of 1058 actions, max excludes first 1056 actions)
> Max place success rate: 0.6666666666666666, at action iteration: 1056. (total of 1059 actions, max excludes first 1056 actions)
> Max action efficiency: 0.0, at action iteration: 1056. (total of 1059 actions, max excludes first 1056 actions)
> saving trial success rate: /home/ahundt/src/real_good_robot/logs/2020-06-11-01-28-16_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/trial-success-rate.log.csv
> saving grasp success rate: /home/ahundt/src/real_good_robot/logs/2020-06-11-01-28-16_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/grasp-success-rate.log.csv
> saving place success rate: /home/ahundt/src/real_good_robot/logs/2020-06-11-01-28-16_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/place-success-rate.log.csv
> saving action efficiency: /home/ahundt/src/real_good_robot/logs/2020-06-11-01-28-16_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/action-efficiency.log.csv
> saving plot: 2020-06-11-01-28-16_Sim-Rows-Two-Step-Reward-Masked-Testing-Sim-Rows-Two-Step-Reward-Masked-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-11-01-28-16_Sim-Rows-Two-Step-Reward-Masked-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-11-01-28-16_Sim-Rows-Two-Step-Reward-Masked-Testing/best_stats.json
> Choosing a snapshot from the following options:{'trial_success_rate_best_value': 0.30666666666666664, 'trial_success_rate_best_index': 6477, 'grasp_success_rate_best_value': 0.9444444444444444, 'grasp_success_rate_best_index': 18688, 'place_success_rate_best_value': 1.0, 'place_success_rate_best_index': 8939, 'action_efficiency_best_value': 0.108, 'action_efficiency_best_index': 2758}
> Evaluating trial_success_rate_best_value
> Shapshot chosen: /home/ahundt/src/real_good_robot/logs/2020-06-07-14-07-53_Sim-Rows-Two-Step-Reward-Masked-Training/models/snapshot.reinforcement_trial_success_rate_best_value.pth
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-07-14-07-53_Sim-Rows-Two-Step-Reward-Masked-Training/2020-06-11-01-28-16_Sim-Rows-Two-Step-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': -inf, 'trial_success_rate_best_index': None, 'grasp_success_rate_best_value': 1.0, 'grasp_success_rate_best_index': 1056, 'place_success_rate_best_value': 0.6666666666666666, 'place_success_rate_best_index': 1056, 'action_efficiency_best_value': 0.0, 'action_efficiency_best_index': 1056}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-07-14-07-53_Sim-Rows-Two-Step-Reward-Masked-Training
> Training results:
> {'trial_success_rate_best_value': 0.30666666666666664, 'trial_success_rate_best_index': 6477, 'grasp_success_rate_best_value': 0.9444444444444444, 'grasp_success_rate_best_index': 18688, 'place_success_rate_best_value': 1.0, 'place_success_rate_best_index': 8939, 'action_efficiency_best_value': 0.108, 'action_efficiency_best_index': 2758}
> Trial logging complete: 101 --------------------------------------------------------------
> STACK: trial: 101 actions/partial: 3.890909090909091 actions/full stack: 39.629629629629626 (lower is better) Grasp Count: 6, grasp success rate: 1.0 place_on_stack_rate: 45.833333333333336 place_attempts: 6 partial_stack_successes: 275 stack_successes: 27 trial_success_rate: 0.26732673267326734 stack
> goal: [1 3] current_height: 2
> trial_complete_indices: [ 8. 19. 27. 41. 50. 56. 59. 69. 77. 84. 98. 107.
> 124. 134. 146. 151. 164. 167. 191. 198. 213. 225. 233. 244.
> 253. 264. 273. 285. 308. 320. 333. 341. 343. 352. 361. 372.
> 397. 408. 419. 429. 435. 445. 448. 455. 465. 480. 492. 509.
> 519. 527. 536. 545. 558. 570. 573. 595. 600. 611. 622. 642.
> 649. 666. 677. 684. 692. 700. 703. 717. 723. 729. 740. 752.
> 760. 771. 776. 791. 802. 813. 819. 828. 840. 853. 864. 877.
> 886. 896. 901. 910. 922. 935. 945. 957. 959. 1001. 1008. 1019.
> 1027. 1036. 1046. 1056. 1069.]
> Max trial success rate: 0.27, at action iteration: 1066. (total of 1068 actions, max excludes first 1066 actions)
> /home/ahundt/src/real_good_robot/plot.py:136: RuntimeWarning: invalid value encountered in double_scalars
> success_rate[i] = float(successes.sum()) / float(grasp_count) if grasp_count > 0 else 0.0
> Max grasp success rate: 1.0, at action iteration: 1066. (total of 1068 actions, max excludes first 1066 actions)
> Max place success rate: 1.0, at action iteration: 1066. (total of 1069 actions, max excludes first 1066 actions)
> Max action efficiency: 0.0, at action iteration: 1066. (total of 1069 actions, max excludes first 1066 actions)
> saving trial success rate: /home/ahundt/src/real_good_robot/logs/2020-06-11-10-22-05_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/trial-success-rate.log.csv
> saving grasp success rate: /home/ahundt/src/real_good_robot/logs/2020-06-11-10-22-05_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/grasp-success-rate.log.csv
> saving place success rate: /home/ahundt/src/real_good_robot/logs/2020-06-11-10-22-05_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/place-success-rate.log.csv
> saving action efficiency: /home/ahundt/src/real_good_robot/logs/2020-06-11-10-22-05_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/action-efficiency.log.csv
> saving plot: 2020-06-11-10-22-05_Sim-Rows-Two-Step-Reward-Masked-Testing-Sim-Rows-Two-Step-Reward-Masked-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-11-10-22-05_Sim-Rows-Two-Step-Reward-Masked-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-11-10-22-05_Sim-Rows-Two-Step-Reward-Masked-Testing/best_stats.json
> Choosing a snapshot from the following options:{'action_efficiency_best_index': 2758, 'action_efficiency_best_value': 0.108, 'grasp_success_rate_best_index': 18688, 'grasp_success_rate_best_value': 0.9444444444444444, 'place_success_rate_best_index': 8939, 'place_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 6477, 'trial_success_rate_best_value': 0.30666666666666664}
> Evaluating trial_success_rate_best_value
> Shapshot chosen: /home/ahundt/src/real_good_robot/logs/2020-06-07-14-07-53_Sim-Rows-Two-Step-Reward-Masked-Training/models/snapshot.reinforcement_trial_success_rate_best_value.pth
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-07-14-07-53_Sim-Rows-Two-Step-Reward-Masked-Training/2020-06-11-10-22-05_Sim-Rows-Two-Step-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 0.27, 'trial_success_rate_best_index': 1066, 'grasp_success_rate_best_value': 1.0, 'grasp_success_rate_best_index': 1066, 'place_success_rate_best_value': 1.0, 'place_success_rate_best_index': 1066, 'action_efficiency_best_value': 0.0, 'action_efficiency_best_index': 1066}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-07-14-07-53_Sim-Rows-Two-Step-Reward-Masked-Training
> Training results:
> {'action_efficiency_best_index': 2758, 'action_efficiency_best_value': 0.108, 'grasp_success_rate_best_index': 18688, 'grasp_success_rate_best_value': 0.9444444444444444, 'place_success_rate_best_index': 8939, 'place_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 6477, 'trial_success_rate_best_value': 0.30666666666666664}
manual action efficiency calculation: 100 trials * 6 ideal actions per trial / 1069 actions = 0.561
SIM STACK - Task Progress aka progress only - RANDOM ACTIONS - REWARD SCHEDULE 0.1, 1, 1 - workstation named spot 2020-06-08
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="2" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --check_z_height --tcp_port 19999 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-08-17-36-39_Sim-Stack-Two-Step-Reward-Training
Commit: 36a0c6a8cfd6c0d8a087f0b647814575054faedd release tag:v0.16.3
GPU 2, Tab 2, port 19990, left v-rep window, v-rep tab 9
> First model test result (second model pending)
> ± '/home/ahundt/src/real_good_robot/logs/2020-06-08-17-36-39_Sim-Stack-Two-Step-Reward-Training/2020-06-12-03-47-07_Sim-Stack-Two-Step-Reward-Testing/best_stats.json'
> {"action_efficiency_best_index": 1542, "action_efficiency_best_value": 0.38132295719844356, "grasp_success_rate_best_index": 1542, "grasp_success_rate_best_value": 0.7957992998833139, "place_success_rate_best_index": 1544, "place_success_rate_best_value": 0.7259475218658892, "trial_success_rate_best_index": 1542, "trial_success_rate_best_value": 0.98}
> Trial logging complete: 101 --------------------------------------------------------------
> check_stack() stack_height: 1.0231710310697297 stack matches current goal: False partial_stack_success: False Does the code think a reset is needed: False
> STACK: trial: 101 actions/partial: 3.937125748502994 actions/full stack: 13.01980198019802 (lower is better) Grasp Count: 735, grasp success rate: 0.7877551020408163 place_on_stack_rate: 0.5768566493955095 place_attempts: 579 partial_stack_successes: 334 stack_successes: 101 trial_success_rate: 1.0 stack goal: None current_height: 1.0231710310697297
> trial_complete_indices: [ 6. 20. 46. 50. 63. 67. 73. 87. 99. 111. 117. 152.
> 158. 166. 181. 187. 197. 205. 209. 219. 232. 242. 250. 265.
> 269. 278. 285. 293. 320. 326. 345. 355. 361. 373. 399. 406.
> 452. 466. 493. 499. 507. 515. 532. 540. 546. 550. 557. 565.
> 571. 577. 593. 601. 610. 618. 639. 643. 657. 664. 672. 678.
> 700. 707. 723. 734. 750. 761. 770. 776. 813. 819. 823. 874.
> 900. 910. 915. 928. 964. 968. 980. 988. 1014. 1023. 1027. 1035.
> 1039. 1070. 1098. 1113. 1127. 1138. 1148. 1168. 1184. 1194. 1200. 1209.
> 1225. 1237. 1261. 1305. 1314.]
> Max trial success rate: 1.0, at action iteration: 1311. (total of 1313 actions, max excludes first 1311 actions)
> Max grasp success rate: 0.7885402455661664, at action iteration: 1311. (total of 1313 actions, max excludes first 1311 actions)
> Max place success rate: 0.7461139896373057, at action iteration: 1311. (total of 1314 actions, max excludes first 1311 actions)
> Max action efficiency: 0.4622425629290618, at action iteration: 1313. (total of 1314 actions, max excludes first 1311 actions)
> saving trial success rate: /home/ahundt/src/real_good_robot/logs/2020-06-12-10-30-14_Sim-Stack-Two-Step-Reward-Testing/transitions/trial-success-rate.log.csv
> saving grasp success rate: /home/ahundt/src/real_good_robot/logs/2020-06-12-10-30-14_Sim-Stack-Two-Step-Reward-Testing/transitions/grasp-success-rate.log.csv
> saving place success rate: /home/ahundt/src/real_good_robot/logs/2020-06-12-10-30-14_Sim-Stack-Two-Step-Reward-Testing/transitions/place-success-rate.log.csv
> saving action efficiency: /home/ahundt/src/real_good_robot/logs/2020-06-12-10-30-14_Sim-Stack-Two-Step-Reward-Testing/transitions/action-efficiency.log.csv
> saving plot: 2020-06-12-10-30-14_Sim-Stack-Two-Step-Reward-Testing-Sim-Stack-Two-Step-Reward-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-12-10-30-14_Sim-Stack-Two-Step-Reward-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-12-10-30-14_Sim-Stack-Two-Step-Reward-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-08-17-36-39_Sim-Stack-Two-Step-Reward-Training/2020-06-12-10-30-14_Sim-Stack-Two-Step-Reward-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 1311, 'grasp_success_rate_best_value': 0.7885402455661664, 'grasp_success_rate_best_index': 1311, 'place_success_rate_best_value': 0.7461139896373057, 'place_success_rate_best_index': 1311, 'action_efficiency_best_value': 0.4622425629290618, 'action_efficiency_best_index': 1313}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-08-17-36-39_Sim-Stack-Two-Step-Reward-Training
> Training results:
> {'trial_success_rate_best_value': 0.8627450980392157, 'trial_success_rate_best_index': 17335, 'grasp_success_rate_best_value': 0.8943396226415095, 'grasp_success_rate_best_index': 18852, 'place_success_rate_best_value': 0.8398268398268398, 'place_success_rate_best_index': 17013, 'action_efficiency_best_value': 0.624, 'action_efficiency_best_index': 17020}
Resumed run
------------
RESUME pass 1 run (formerly gpu 1) on gpu 3, see /home/ahundt/src/real_good_robot/logs/2020-05-28-12-10-18_Sim-Rows-SPOT-Trial-Reward-Training near the top of this file
PASS 4
========================================
PUSHING AND GRASPING WITH TRIAL REWARD & SAVE ALL MODELS ACCORDING TO BEST STATS - 2020-06-12
--------------------------------------
export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/toys --num_obj 10 --push_rewards --experience_replay --explore_rate_decay --tcp_port 19990 --common_sense --trial_reward --future_reward_discount 0.65 --random_actions --max_train_actions 5000
Commit: 60c816e2539f9b105f622132d1a6e22dc572caff release tag:v0.16.4
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-12-00-36-00_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training
GPU 0, Tab 0, port 19990, left v-rep window, v-rep tab 7
RESUME: ± export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/toys --num_obj 10 --push_rewards --experience_replay --explore_rate_decay --tcp_port 19990 --common_sense --trial_reward --future_reward_discount 0.65 --random_actions --max_train_actions 5000 --resume /home/ahundt/src/real_good_robot/logs/2020-06-12-00-36-00_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training
RESUME COMMIT: d0294ba8f84feea20ade3f3ec7ba9ba96a6b9371 (plotting crash bugfix)
± '/home/ahundt/src/real_good_robot/logs/2020-06-12-00-36-00_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training/2020-06-13-07-52-11_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Testing/best_stats.json'
Random testing:
{"grasp_action_efficiency_best_index": 1562, "grasp_action_efficiency_best_value": 0.6165172855313701, "grasp_success_rate_best_index": 1562, "grasp_success_rate_best_value": 0.7351145038167939, "trial_success_rate_best_index": 1562, "trial_success_rate_best_value": 0.98989898989899}
testing challenging arrangements:±
'/home/ahundt/src/real_good_robot/logs/2020-06-12-00-36-00_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training/2020-06-13-15-04-48_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/best_stats.json'
{"grasp_action_efficiency_best_index": 1442, "grasp_action_efficiency_best_value": 0.3393476752255378, "grasp_success_rate_best_index": 1442, "grasp_success_rate_best_value": 0.443336355394379, "senarios_100_percent_complete": 6, "trial_success_rate_best_index": 1441, "trial_success_rate_best_value": 0.9541284403669725}
PUSHING AND GRASPING WITH TASK PROGRESS & SAVE ALL MODELS ACCORDING TO BEST STATS - 2020-06-13
--------------------------------------
export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/toys --num_obj 10 --push_rewards --experience_replay --explore_rate_decay --tcp_port 19990 --common_sense --future_reward_discount 0.65 --random_actions --max_train_actions 5000
Commit: 60c816e2539f9b105f622132d1a6e22dc572caff release tag:v0.16.4
Creating data logging session:
GPU 0, Tab 0, port 19990, left v-rep window, v-rep tab 7
Parameter sensitivity experiment - SIM ROW - Task Progress SPOT-Q MASKED - REWARD SCHEDULE 0.5, 1, 1 - workstation named spot 2020-06-12
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="1" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --check_row --tcp_port 19998 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-11-23-41-01_Sim-Rows-Two-Step-Reward-Masked-Training
Commit: ff54ed3ae8b700b12433897e680362675cd31013 release tag:push_r_weight_0.5_v1
GPU 1, Tab 1, port 19998, center left v-rep window, v-rep tab 8
ANY OBJECT SIM STACK - Trial Reward SPOT-Q MASKED - RANDOM ACTIONS - REWARD SCHEDULE 0.1, 1, 1 - EFFICIENTNET 1 dilation - workstation named spot 2020-06-03
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="2" && python3 main.py --is_sim --obj_mesh_dir objects/toys --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --check_z_height --tcp_port 19999 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --trial_reward --common_sense --nn efficientnet --num_dilation 1
Creating data logging session:
Commit: 60c816e2539f9b105f622132d1a6e22dc572caff release tag:v0.16.4
GPU 2, Tab 2, port 19999, right v-rep window, v-rep tab 10
Resumed run
------------
RESUME pass 1 run (formerly gpu 0) on gpu 3, see /home/ahundt/src/real_good_robot/logs/2020-05-28-12-09-32_Sim-Stack-SPOT-Trial-Reward-Training near the top of this file
PASS 5 - efficientnet
=======
PUSHING AND GRASPING WITH TRIAL REWARD & SAVE ALL MODELS ACCORDING TO BEST STATS - EFFICIENTNET 1 dilation - 2020-06-26
--------------------------------------
export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/toys --num_obj 10 --push_rewards --experience_replay --explore_rate_decay --tcp_port 19990 --common_sense --trial_reward --future_reward_discount 0.65 --random_actions --max_train_actions 5000 --nn efficientnet --num_dilation 1
Commit: cca5887217884d862167edbf31ffaf4d1d21a863 release tag:v0.16.5
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-10-40_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training
GPU 0, Tab 0, port 19990, left v-rep window, v-rep tab 7
> Trial logging complete: 110 --------------------------------------------------------------
> Running two step backprop()
> Primitive confidence scores: 0.746463 (push), 1.726852 (grasp)
> Action: grasp at (4, 92, 83)
> Training loss: 0.057178
> prev_height: 0.0 max_z: 0.05111149809658745 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> Grasp Count: 986, grasp success rate: 0.513184584178499
> trial_complete_indices: [ 6. 13. 17. 28. 39. 54. 61. 70. 79. 85. 96. 106. 118. 132.
> 146. 156. 163. 170. 178. 187. 192. 199. 207. 210. 212. 218. 222. 229.
> 233. 240. 250. 264. 279. 286. 296. 308. 322. 337. 349. 360. 370. 377.
> 390. 398. 404. 415. 424. 430. 438. 444. 454. 463. 473. 481. 487. 495.
> 503. 508. 516. 523. 530. 535. 541. 546. 551. 556. 562. 566. 575. 583.
> 593. 606. 615. 628. 641. 651. 660. 670. 682. 688. 698. 714. 730. 744.
> 756. 767. 775. 781. 805. 813. 821. 827. 836. 847. 854. 860. 865. 871.
> 880. 889. 897. 909. 916. 924. 936. 942. 953. 965. 979. 990.]
> Max trial success rate: 1.0, at action iteration: 987. (total of 989 actions, max excludes first 987 actions)
> max trial successes: 110.0
> individual_arrangement_trial_success_rate: [1. 1. 0.9 1. 1. 1. 1. 1. 1. 1. 0.9]
> senarios_100_percent_complete: 9
> Max grasp success rate: 0.513733468972533, at action iteration: 987. (total of 989 actions, max excludes first 987 actions)
> Max grasp action efficiency: 0.5116514690982776, at action iteration: 987. (total of 990 actions, max excludes first 987 actions)
> saving trial success rate: /home/ahundt/src/real_good_robot/logs/2020-06-27-18-18-07_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/transitions/trial-success-rate.log.csv
> saving grasp success rate: /home/ahundt/src/real_good_robot/logs/2020-06-27-18-18-07_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/transitions/grasp-success-rate.log.csv
> saving action efficiency: /home/ahundt/src/real_good_robot/logs/2020-06-27-18-18-07_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/transitions/action-efficiency.log.csv
> saving plot: 2020-06-27-18-18-07_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements-Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-27-18-18-07_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-27-18-18-07_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/best_stats.json
> Challenging Arrangements Preset Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-10-40_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training/2020-06-27-18-18-07_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements
> Challenging Arrangements Preset Testing results:
> {'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 987, 'senarios_100_percent_complete': 9, 'grasp_success_rate_best_value': 0.513733468972533, 'grasp_success_rate_best_index': 987, 'grasp_action_efficiency_best_value': 0.5116514690982776, 'grasp_action_efficiency_best_index': 987}
> Choosing a snapshot from the following options:{'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 675, 'grasp_success_rate_best_value': 0.8714285714285714, 'grasp_success_rate_best_index': 4705, 'grasp_action_efficiency_best_value': 0.854, 'grasp_action_efficiency_best_index': 4705}
> Evaluating trial_success_rate_best_value
> Shapshot chosen: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-10-40_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training/models/snapshot.reinforcement_trial_success_rate_best_value.pth
> Challenging Arrangements Preset Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-10-40_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training/2020-06-27-18-18-07_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements
> Challenging Arrangements Preset Testing results:
> {'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 987, 'senarios_100_percent_complete': 9, 'grasp_success_rate_best_value': 0.513733468972533, 'grasp_success_rate_best_index': 987, 'grasp_action_efficiency_best_value': 0.5116514690982776, 'grasp_action_efficiency_best_index': 987}
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-10-40_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training/2020-06-27-13-00-19_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 1133, 'grasp_success_rate_best_value': 0.8499558693733451, 'grasp_success_rate_best_index': 1133, 'grasp_action_efficiency_best_value': 0.8499558693733451, 'grasp_action_efficiency_best_index': 1133}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-10-40_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training
> Training results:
> {'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 675, 'grasp_success_rate_best_value': 0.8714285714285714, 'grasp_success_rate_best_index': 4705, 'grasp_action_efficiency_best_value': 0.854, 'grasp_action_efficiency_best_index': 4705}
PUSHING AND GRASPING WITH TRIAL REWARD & SAVE ALL MODELS ACCORDING TO BEST STATS - EFFICIENTNET 1 dilation - 2020-06-28
--------------------------------------
export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/toys --num_obj 10 --push_rewards --experience_replay --explore_rate_decay --tcp_port 19990 --common_sense --trial_reward --future_reward_discount 0.65 --random_actions --max_train_actions 5000 --nn efficientnet --num_dilation 1
Commit: cca5887217884d862167edbf31ffaf4d1d21a863 release tag:v0.16.5
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-28-10-45-13_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training
GPU 0, Tab 0, port 19990, left v-rep window, v-rep tab 7
> Trial logging complete: 110 --------------------------------------------------------------
> Running two step backprop()
> Primitive confidence scores: 1.113376 (push), 1.805528 (grasp)
> Action: grasp at (13, 139, 115)
> Training loss: 0.197310
> Executing: grasp at (-0.494000, 0.054000, 0.050216) orientation: 5.105088
> gripper position: 0.0528782494366169
> gripper position: 0.034795910120010376
> gripper position: 0.033892154693603516
> gripper position: 0.03291165828704834
> Grasp successful: False
> prev_height: 0.0 max_z: 0.0667136022755871 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> Grasp Count: 852, grasp success rate: 0.5880281690140845
> trial_complete_indices: [ 8. 18. 33. 41. 48. 58. 69. 84. 95. 101. 108. 120. 127. 134.
> 140. 152. 160. 169. 179. 191. 196. 202. 208. 214. 221. 227. 232. 236.
> 241. 255. 268. 275. 286. 301. 316. 328. 343. 358. 368. 378. 384. 391.
> 400. 408. 420. 426. 434. 441. 449. 455. 464. 472. 481. 492. 501. 509.
> 518. 527. 538. 546. 551. 555. 559. 566. 570. 576. 580. 586. 590. 594.
> 606. 615. 624. 632. 640. 647. 655. 662. 670. 682. 696. 709. 721. 733.
> 753. 765. 780. 794. 809. 823. 829. 837. 847. 858. 866. 873. 883. 896.
> 910. 916. 924. 932. 941. 947. 953. 963. 969. 975. 984. 992.]
> Max trial success rate: 1.0, at action iteration: 989. (total of 991 actions, max excludes first 989 actions)
> max trial successes: 110.0
> individual_arrangement_trial_success_rate: [1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 0.9]
> senarios_100_percent_complete: 10
> Max grasp success rate: 0.5882352941176471, at action iteration: 990. (total of 991 actions, max excludes first 989 actions)
> Max grasp action efficiency: 0.506572295247725, at action iteration: 991. (total of 992 actions, max excludes first 989 actions)
> saving trial success rate: /home/ahundt/src/real_good_robot/logs/2020-06-29-13-27-18_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/transitions/trial-success-rate.log.csv
> saving grasp success rate: /home/ahundt/src/real_good_robot/logs/2020-06-29-13-27-18_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/transitions/grasp-success-rate.log.csv
> saving action efficiency: /home/ahundt/src/real_good_robot/logs/2020-06-29-13-27-18_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/transitions/action-efficiency.log.csv
> saving plot: 2020-06-29-13-27-18_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements-Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-29-13-27-18_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-29-13-27-18_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements/best_stats.json
> Challenging Arrangements Preset Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-28-10-45-13_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training/2020-06-29-13-27-18_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements
> Challenging Arrangements Preset Testing results:
> {'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 989, 'senarios_100_percent_complete': 10, 'grasp_success_rate_best_value': 0.5882352941176471, 'grasp_success_rate_best_index': 990, 'grasp_action_efficiency_best_value': 0.506572295247725, 'grasp_action_efficiency_best_index': 991}
> Choosing a snapshot from the following options:{'trial_success_rate_best_value': 1.0232558139534884, 'trial_success_rate_best_index': 4999, 'grasp_success_rate_best_value': 0.8843813387423936, 'grasp_success_rate_best_index': 4768, 'grasp_action_efficiency_best_value': 0.872, 'grasp_action_efficiency_best_index': 4768}
> Evaluating trial_success_rate_best_value
> Shapshot chosen: /home/ahundt/src/real_good_robot/logs/2020-06-28-10-45-13_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training/models/snapshot.reinforcement_trial_success_rate_best_value.pth
> Challenging Arrangements Preset Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-28-10-45-13_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training/2020-06-29-13-27-18_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Challenging-Arrangements
> Challenging Arrangements Preset Testing results:
> {'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 989, 'senarios_100_percent_complete': 10, 'grasp_success_rate_best_value': 0.5882352941176471, 'grasp_success_rate_best_index': 990, 'grasp_action_efficiency_best_value': 0.506572295247725, 'grasp_action_efficiency_best_index': 991}
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-28-10-45-13_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training/2020-06-29-08-15-52_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 1.0, 'trial_success_rate_best_index': 1122, 'grasp_success_rate_best_value': 0.8556053811659193, 'grasp_success_rate_best_index': 1122, 'grasp_action_efficiency_best_value': 0.8502673796791443, 'grasp_action_efficiency_best_index': 1122}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-28-10-45-13_Sim-Push-and-Grasp-SPOT-Trial-Reward-Masked-Training
> Training results:
> {'trial_success_rate_best_value': 1.0232558139534884, 'trial_success_rate_best_index': 4999, 'grasp_success_rate_best_value': 0.8843813387423936, 'grasp_success_rate_best_index': 4768, 'grasp_action_efficiency_best_value': 0.872, 'grasp_action_efficiency_best_index': 4768}
ANY OBJECT SIM STACK - Task Progress SPOT-Q MASKED - RANDOM ACTIONS - REWARD SCHEDULE 0.1, 1, 1 - EFFICIENTNET 1 dilation - workstation named spot 2020-06-26
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="1" && python3 main.py --is_sim --obj_mesh_dir objects/toys --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --check_z_height --tcp_port 19998 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense --nn efficientnet --num_dilation 1
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-13-25_Sim-Stack-Two-Step-Reward-Masked-Training
Commit: cca5887217884d862167edbf31ffaf4d1d21a863 release tag:v0.16.5
GPU 1, Tab 1, port 19998, right center v-rep window, v-rep tab 8
> Trial logging complete: 101 --------------------------------------------------------------
> STACK: trial: 101 actions/partial: 29.161125319693095 actions/full stack: 438.53846153846155 (lower is better) Grasp Count: 6677, grasp success rate: 0.5800509210723379 place_on_stack_rate: 0.1013215859030837 place_attempts: 3859 partial_stack_successes: 391 stack_successes: 26 trial_success_rate: 0.25742574257425743 stack goal: None current_height: 1.3412568889607048
> trial_complete_indices: [ 225. 366. 460. 571. 752. 835. 966. 1122. 1221. 1364.
> 1492. 1545. 1680. 1741. 1829. 1895. 2040. 2189. 2443. 2554.
> 2714. 2753. 2810. 2856. 2985. 3120. 3295. 3416. 3508. 3572.
> 3669. 3732. 3862. 3938. 4024. 4085. 4112. 4175. 4186. 4342.
> 4492. 4677. 4749. 4904. 5083. 5167. 5297. 5412. 5581. 5698.
> 5896. 5922. 6073. 6167. 6269. 6355. 6659. 6714. 6743. 6894.
> 6975. 7053. 7116. 7211. 7342. 7418. 7519. 7549. 7796. 7820.
> 7858. 8041. 8109. 8183. 8428. 8638. 8894. 9005. 9114. 9337.
> 9464. 9686. 9738. 9817. 9961. 10015. 10145. 10193. 10328. 10366.
> 10445. 10548. 10626. 10742. 10764. 10848. 10865. 11031. 11197. 11315.
> 11401.]
> Max trial success rate: 0.26, at action iteration: 11398. (total of 11400 actions, max excludes first 11398 actions)
> Max grasp success rate: 0.5801618219958046, at action iteration: 11398. (total of 11400 actions, max excludes first 11398 actions)
> Max place success rate: 0.5263492063492063, at action iteration: 11398. (total of 11401 actions, max excludes first 11398 actions)
> Max action efficiency: 0.015265836111598525, at action iteration: 11398. (total of 11401 actions, max excludes first 11398 actions)
> saving trial success rate: /home/ahundt/src/real_good_robot/logs/2020-07-02-14-53-00_Sim-Stack-Two-Step-Reward-Masked-Testing/transitions/trial-success-rate.log.csv
> saving grasp success rate: /home/ahundt/src/real_good_robot/logs/2020-07-02-14-53-00_Sim-Stack-Two-Step-Reward-Masked-Testing/transitions/grasp-success-rate.log.csv
> saving place success rate: /home/ahundt/src/real_good_robot/logs/2020-07-02-14-53-00_Sim-Stack-Two-Step-Reward-Masked-Testing/transitions/place-success-rate.log.csv
> saving action efficiency: /home/ahundt/src/real_good_robot/logs/2020-07-02-14-53-00_Sim-Stack-Two-Step-Reward-Masked-Testing/transitions/action-efficiency.log.csv
> saving plot: 2020-07-02-14-53-00_Sim-Stack-Two-Step-Reward-Masked-Testing-Sim-Stack-Two-Step-Reward-Masked-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-07-02-14-53-00_Sim-Stack-Two-Step-Reward-Masked-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-07-02-14-53-00_Sim-Stack-Two-Step-Reward-Masked-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-13-25_Sim-Stack-Two-Step-Reward-Masked-Training/2020-06-30-05-17-35_Sim-Stack-Two-Step-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 0.47, 'trial_success_rate_best_index': 13484, 'grasp_success_rate_best_value': 0.7247529834466829, 'grasp_success_rate_best_index': 13484, 'place_success_rate_best_value': 0.5312719606465214, 'place_success_rate_best_index': 13486, 'action_efficiency_best_value': 0.027588252743992882, 'action_efficiency_best_index': 13486}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-13-25_Sim-Stack-Two-Step-Reward-Masked-Training
> Training results:
> {'trial_success_rate_best_value': 0.25, 'trial_success_rate_best_index': 13316, 'grasp_success_rate_best_value': 0.8694029850746269, 'grasp_success_rate_best_index': 18482, 'place_success_rate_best_value': 0.6090909090909091, 'place_success_rate_best_index': 8898, 'action_efficiency_best_value': 1.164, 'action_efficiency_best_index': 6812}
SIM STACK - Task Progress SPOT-Q MASKED - RANDOM ACTIONS - REWARD SCHEDULE 0.1, 1, 1 - EFFICIENTNET 1 dilation - workstation named spot 2020-06-26
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="2" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --check_z_height --tcp_port 19999 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense --nn efficientnet --num_dilation 1
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-17-08_Sim-Stack-Two-Step-Reward-Masked-Training
Commit: cca5887217884d862167edbf31ffaf4d1d21a863 release tag:v0.16.5
GPU 2, Tab 2, port 19999, right center v-rep window, v-rep tab 9
> Trial logging complete: 101 --------------------------------------------------------------
> prev_height: 0.0 max_z: 0.051130634964013114 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> prev_height: 1.0 max_z: 1.0226126992802622 goal_success: False needed to reset: False max_workspace_height: 0.6 <<<<<<<<<<<
> check_stack() stack_height: 1.0226126992802622 stack matches current goal: False partial_stack_success: False Does the code think a reset is needed: False
> STACK: trial: 101 actions/partial: 3.0235690235690234 actions/full stack: 8.98 (lower is better) Grasp Count: 488, grasp success rate: 0.8422131147540983 place_on_stack_rate: 0.724390243902439 place_attempts: 410 partial_stack_successes: 297 stack_successes: 100 trial_success_rate: 0.9900990099009901 stack goal: None current_height: 1.0226126992802622
> trial_complete_indices: [ 6. 20. 32. 38. 47. 53. 61. 69. 73. 78. 88. 93. 99. 105.
> 113. 119. 125. 137. 145. 149. 155. 159. 171. 186. 194. 198. 203. 214.
> 224. 230. 236. 244. 253. 259. 265. 275. 281. 287. 293. 301. 311. 322.
> 330. 342. 350. 361. 368. 374. 383. 390. 398. 402. 418. 425. 433. 441.
> 445. 453. 457. 471. 479. 485. 492. 502. 550. 559. 575. 581. 587. 595.
> 599. 606. 616. 628. 634. 638. 649. 656. 666. 674. 680. 686. 691. 700.
> 707. 711. 717. 729. 735. 743. 753. 762. 769. 775. 785. 803. 815. 823.
> 880. 889. 897.]
> Max trial success rate: 0.99, at action iteration: 894. (total of 896 actions, max excludes first 894 actions)
> Max grasp success rate: 0.8436213991769548, at action iteration: 894. (total of 896 actions, max excludes first 894 actions)
> Max place success rate: 0.8484107579462102, at action iteration: 894. (total of 897 actions, max excludes first 894 actions)
> Max action efficiency: 0.6711409395973155, at action iteration: 896. (total of 897 actions, max excludes first 894 actions)
> saving trial success rate: /home/ahundt/src/real_good_robot/logs/2020-06-30-10-36-48_Sim-Stack-Two-Step-Reward-Masked-Testing/transitions/trial-success-rate.log.csv
> saving grasp success rate: /home/ahundt/src/real_good_robot/logs/2020-06-30-10-36-48_Sim-Stack-Two-Step-Reward-Masked-Testing/transitions/grasp-success-rate.log.csv
> saving place success rate: /home/ahundt/src/real_good_robot/logs/2020-06-30-10-36-48_Sim-Stack-Two-Step-Reward-Masked-Testing/transitions/place-success-rate.log.csv
> saving action efficiency: /home/ahundt/src/real_good_robot/logs/2020-06-30-10-36-48_Sim-Stack-Two-Step-Reward-Masked-Testing/transitions/action-efficiency.log.csv
> saving plot: 2020-06-30-10-36-48_Sim-Stack-Two-Step-Reward-Masked-Testing-Sim-Stack-Two-Step-Reward-Masked-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-30-10-36-48_Sim-Stack-Two-Step-Reward-Masked-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-30-10-36-48_Sim-Stack-Two-Step-Reward-Masked-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-17-08_Sim-Stack-Two-Step-Reward-Masked-Training/2020-06-30-06-33-03_Sim-Stack-Two-Step-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 0.99, 'trial_success_rate_best_index': 823, 'grasp_success_rate_best_value': 0.9133489461358314, 'grasp_success_rate_best_index': 823, 'place_success_rate_best_value': 0.8513853904282116, 'place_success_rate_best_index': 823, 'action_efficiency_best_value': 0.7290400972053463, 'action_efficiency_best_index': 825}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-17-08_Sim-Stack-Two-Step-Reward-Masked-Training
> Training results:
> {'trial_success_rate_best_value': 0.9054054054054054, 'trial_success_rate_best_index': 15101, 'grasp_success_rate_best_value': 0.9683794466403162, 'grasp_success_rate_best_index': 12884, 'place_success_rate_best_value': 0.9267241379310345, 'place_success_rate_best_index': 13640, 'action_efficiency_best_value': 0.864, 'action_efficiency_best_index': 13081}
SIM ROW - Task Progress SPOT-Q MASKED - REWARD SCHEDULE 0.1, 1, 1 - EFFICIENTNET 1 dilation - workstation named spot 2020-06-26
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="3" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --check_row --tcp_port 20000 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense --nn efficientnet --num_dilation 1
Creating data logging session: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-18-44_Sim-Rows-Two-Step-Reward-Masked-Training
Commit: cca5887217884d862167edbf31ffaf4d1d21a863 release tag:v0.16.5
GPU 3, Tab 3, port 20000, right v-rep window, v-rep tab 10
> Trial logging complete: 101 --------------------------------------------------------------
> prev_height: 0.0 max_z: 0.05110832959176721 goal_success: True needed to reset: False max_workspace_height: -0.02 <<<<<<<<<<<
> check_row: False | row_size: 1 | blocks: []
> check_stack() stack_height: 1 stack matches current goal: True partial_stack_success: False Does the code think a reset is needed: False
> STACK: trial: 101 actions/partial: 5.036144578313253 actions/full stack: 12.666666666666666 (lower is better) Grasp Count: 677, grasp success rate: 0.8552437223042836 place_on_stack_rate: 0.43154246100519933 place_attempts: 577 partial_stack_successes: 249 stack_successes: 99 trial_success_rate: 0.9801980198019802 stack goal: [2 1] current_height: 1
> trial_complete_indices: [ 4. 8. 14. 22. 34. 51. 55. 59. 63. 78. 88. 98.
> 105. 139. 143. 164. 166. 188. 195. 214. 218. 224. 232. 237.
> 276. 282. 287. 307. 340. 344. 352. 364. 368. 372. 390. 398.
> 402. 419. 421. 425. 450. 460. 468. 490. 494. 502. 508. 522.
> 524. 529. 541. 547. 661. 713. 717. 754. 774. 782. 804. 816.
> 822. 829. 834. 838. 842. 847. 849. 856. 885. 889. 894. 932.
> 934. 953. 983. 1006. 1010. 1018. 1030. 1040. 1044. 1078. 1110. 1118.
> 1136. 1140. 1151. 1153. 1161. 1167. 1183. 1189. 1195. 1214. 1218. 1220.
> 1226. 1236. 1240. 1244. 1253.]
> Max trial success rate: 0.98, at action iteration: 1250. (total of 1252 actions, max excludes first 1250 actions)
> Max grasp success rate: 0.8562962962962963, at action iteration: 1250. (total of 1252 actions, max excludes first 1250 actions)
> Max place success rate: 0.7013888888888888, at action iteration: 1250. (total of 1253 actions, max excludes first 1250 actions)
> Max action efficiency: 0.4896, at action iteration: 1252. (total of 1253 actions, max excludes first 1250 actions)
> saving trial success rate: /home/ahundt/src/real_good_robot/logs/2020-06-30-14-25-16_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/trial-success-rate.log.csv
> saving grasp success rate: /home/ahundt/src/real_good_robot/logs/2020-06-30-14-25-16_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/grasp-success-rate.log.csv
> saving place success rate: /home/ahundt/src/real_good_robot/logs/2020-06-30-14-25-16_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/place-success-rate.log.csv
> saving action efficiency: /home/ahundt/src/real_good_robot/logs/2020-06-30-14-25-16_Sim-Rows-Two-Step-Reward-Masked-Testing/transitions/action-efficiency.log.csv
> saving plot: 2020-06-30-14-25-16_Sim-Rows-Two-Step-Reward-Masked-Testing-Sim-Rows-Two-Step-Reward-Masked-Testing_success_plot.png
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-30-14-25-16_Sim-Rows-Two-Step-Reward-Masked-Testing/data/best_stats.json
> saving best stats to: /home/ahundt/src/real_good_robot/logs/2020-06-30-14-25-16_Sim-Rows-Two-Step-Reward-Masked-Testing/best_stats.json
> Random Testing Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-18-44_Sim-Rows-Two-Step-Reward-Masked-Training/2020-06-30-14-25-16_Sim-Rows-Two-Step-Reward-Masked-Testing
> Random Testing results:
> {'trial_success_rate_best_value': 0.98, 'trial_success_rate_best_index': 1250, 'grasp_success_rate_best_value': 0.8562962962962963, 'grasp_success_rate_best_index': 1250, 'place_success_rate_best_value': 0.7013888888888888, 'place_success_rate_best_index': 1250, 'action_efficiency_best_value': 0.4896, 'action_efficiency_best_index': 1252}
> Training Complete! Dir: /home/ahundt/src/real_good_robot/logs/2020-06-26-15-18-44_Sim-Rows-Two-Step-Reward-Masked-Training
> Training results:
> {'trial_success_rate_best_value': 0.5869565217391305, 'trial_success_rate_best_index': 13103, 'grasp_success_rate_best_value': 0.7921146953405018, 'grasp_success_rate_best_index': 15530, 'place_success_rate_best_value': 0.7102803738317757, 'place_success_rate_best_index': 15717, 'action_efficiency_best_value': 0.708, 'action_efficiency_best_index': 15733}
SIM STACK - Task Progress SPOT-Q MASKED - RANDOM ACTIONS - REWARD SCHEDULE 0.1, 1, 1 - EFFICIENTNET 1 dilation - workstation named spot 2020-06-30
----------------------------------------------------------------------------------------
export CUDA_VISIBLE_DEVICES="2" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 8 --push_rewards --experience_replay --explore_rate_decay --check_z_height --tcp_port 19999 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense --nn efficientnet --num_dilation 1