Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

native Cuda kernel for tf.dynamic_stitch op #7764

Closed
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
666 commits
Select commit Hold shift + click to select a range
5e9a361
Update ops-related pbtxt files.
tensorflower-gardener Mar 2, 2017
8cac382
Merge pull request #7985 from ClarkZinzow/ClarkZinzow-patch-1
caisq Mar 2, 2017
38c9f46
Move build_*_serving_input_fn to core as build_*_serving_input_receiv…
davidsoergel Mar 2, 2017
2520eae
tfdbg: fix source_utils on Windows
caisq Mar 2, 2017
df95a02
Exclude a few files from doc generation.
tensorflower-gardener Mar 2, 2017
1cdf3b3
A simple GPU build script.
gunan Mar 2, 2017
75e4512
Automated rollback of change 148905332
tensorflower-gardener Mar 2, 2017
e13134f
[ios camera app] Unlock pixelBuffer after prediction
tomergafner Mar 2, 2017
860056b
[ios camera app] replace deprecated stringWithCString
tomergafner Mar 2, 2017
a7c58c8
Make tf.contrib.memory_stats.MaxBytesInUse accessible.
Mar 2, 2017
ceb293f
small grammar/typo edits
yufengg Mar 2, 2017
cac52fd
fixed the filter position cacluating for the VALID padding mode
Mar 2, 2017
f2e3d41
All distributions: Use init arg "name" for the name, rather than nam…
langmore Mar 2, 2017
ceb0462
Fixed broken link to Community
calpa Mar 2, 2017
1a26987
Adds model_dir into (contrib) RunConfig and validates it in Estimator…
Mar 2, 2017
a0db266
Clean up matmul test suites and make sure that complex tests set the …
tensorflower-gardener Mar 2, 2017
f9694e8
Fix metagemm quantization offsets in matmul.
tensorflower-gardener Mar 2, 2017
ab49d4c
tfdbg CLI: a bug fix and a UI improvement
caisq Mar 2, 2017
57d97f4
Improved wording of MonitoredSession preemption error logging.
tensorflower-gardener Mar 2, 2017
e95219b
Reorder service.h methods to group TransferTo* methods.
saeta Mar 2, 2017
87d3f8f
Fix local_test.sh to respect --num_workers and --num_parameter_servers
cberner Mar 2, 2017
87ce73a
Add descriptions for Conv2D "input", "filter", and "output" parameters.
Mar 2, 2017
02e53cf
Tracking lookup table memory usage.
Mar 2, 2017
b8feac5
Retrying FileExists errors in case of errors to overcome intermittent…
tensorflower-gardener Mar 2, 2017
3190b79
When a list of strings is passed to one of the read_batch_* methods, …
tensorflower-gardener Mar 2, 2017
c35d948
Update ops-related pbtxt files.
tensorflower-gardener Mar 2, 2017
c28b1ec
Go: Update generated wrapper functions for TensorFlow ops.
tensorflower-gardener Mar 2, 2017
356e599
Print whether a node is_dead in the executor log statement.
Mar 2, 2017
54c0ff4
Add BenchmarkType to TestResults proto and fix logging of C++ microbe…
ebrevdo Mar 2, 2017
a915636
Implement sharding by input row for depthwiseconv2d, allowing multith…
andrewharp Mar 2, 2017
2f9c1d2
Improve speed of depthwise conv backward on GPU.
tensorflower-gardener Mar 2, 2017
fee34c1
Tracking memory usage of queues.
Mar 2, 2017
379560b
[TF:XLA] Reduce sequential memory usage via better ordering and simul…
tensorflower-gardener Mar 2, 2017
37b4005
Fix typos and inconsistencies in layers docstrings.
fchollet Mar 2, 2017
83860e2
Added interfaces for the messages in contrib/../graph_explorer.proto.
vdsilva Mar 2, 2017
2fa3f68
Allow setting name on accuracy op
cancan101 Mar 2, 2017
3fe917b
Fix bug in matmul gradients for complex types.
tensorflower-gardener Mar 2, 2017
06fd46a
Ensure that QuantizeAndDequantizeV2Op doesn't alter modify input tens…
Mar 3, 2017
509898b
Error out if a subclass of Estimator overrides a member of Estimator.
ispirmustafa Mar 3, 2017
882e75d
Add tfcompile target triple for android_x86.
tensorflower-gardener Mar 3, 2017
d2d5f19
Upon any HTTP error, clear the response buffer.
Mar 3, 2017
112a534
[XLA:GPU] Cache GPU substreams across executions
tensorflower-gardener Mar 3, 2017
dc6353d
Merge pull request #8001 from yufengg/patch-1
caisq Mar 3, 2017
e065b30
[StreamExecutor] Minor comment cleanups.
Mar 3, 2017
0119469
[XLA] [StreamExecutor] Tune GEMMs when possible.
Mar 3, 2017
d8e5efa
Fix the cmake failure: ambiguous symbol
Mar 3, 2017
0e578f1
Android: add support for building libtensorflow_inference.so with mak…
andrewharp Mar 3, 2017
00b79b4
Avoid creating a tensorflow::Status::OK() object and destroying it on
jeffreyadean Mar 3, 2017
3035ee4
Moves build.gradle to use gradle plugin 2.2
markmcd Mar 3, 2017
26df62c
Merge pull request #7981 from Conchylicultor/master
caisq Mar 3, 2017
afd0244
Interal change.
tensorflower-gardener Mar 3, 2017
4ff4702
Fully qualify symbols to prevent clashes in directories named 'tensor…
tensorflower-gardener Mar 3, 2017
13521e1
Windows: Fix error in release script for C library
asimshankar Mar 3, 2017
c91e568
Create contrib folder for new boosted_trees library.
tensorflower-gardener Mar 3, 2017
1f09d0c
Merge pull request #8018 from calpa/patch-1
caisq Mar 3, 2017
2299ffe
Update build-file to expose framework/types.proto temporarily.
tensorflower-gardener Mar 3, 2017
936e593
Does pandas import check in each caller file directly to avoid flaky …
Mar 3, 2017
f864d89
Doc generator refactoring.
tensorflower-gardener Mar 3, 2017
c54417f
Increase the number of retries to 10.
rinugun Mar 3, 2017
20af1b7
Fix python paths in windows cmake build.
tensorflower-gardener Mar 3, 2017
d495250
Merge pull request #7724 from malmaud/control_slot
caisq Mar 3, 2017
51eef31
Merge
Mar 3, 2017
88c9fb0
Restore the adding_an_op code examples that used to live under
tensorflower-gardener Mar 3, 2017
067cba5
Merge pull request #8062 from jhseu/branch_149132162
jhseu Mar 3, 2017
3b59692
Factor-out dryrun functionalities from graph_transfer_utils to remote…
tensorflower-gardener Mar 3, 2017
a868af7
Retry failures when requesting auth token from GCE metadata server.
rinugun Feb 14, 2017
9695077
Change pip.sh to take Bazel flags directly so AVX packages are built …
jhseu Feb 23, 2017
1ded480
Add a non-debug avx2 fma build
Feb 24, 2017
d4fd3b1
Merge
rinugun Mar 3, 2017
aee4914
Retrying FileExists errors in case of errors to overcome intermittent…
tensorflower-gardener Mar 2, 2017
cbb612e
Upon any HTTP error, clear the response buffer.
Mar 3, 2017
4521031
In ShapeRefiner::ExtractConstantSubgraph, don't include
tensorflower-gardener Mar 3, 2017
2924e92
Fix merge
Mar 3, 2017
3a0924b
Fix the error message returned from _init_from_args function in Queue…
tensorflower-gardener Mar 3, 2017
eaa668e
[TF:XLA] Implement SparseSoftmaxCrossEntropyWithLogits.
hawkinsp Mar 3, 2017
1dc89c1
[TF:XLA] Clarify order of operations for ReduceWindow and SelectAndSc…
hawkinsp Mar 3, 2017
b177e37
Add the graphdef version to InferenceContext and to ShapeRefiner::Add…
tensorflower-gardener Mar 3, 2017
3f49b5c
Add a script to run xla tests.
gunan Mar 3, 2017
60e7360
Clipping into [0.0, 1.0] before converting back to original data type…
tensorflower-gardener Mar 3, 2017
571c40a
Merge commit for internal changes
Mar 3, 2017
16a4bc2
Make bidirectional_dynamic_rnn sequence_length argument optional (Fix…
ebrevdo Mar 3, 2017
dd2be3c
Drop "x","y" args in favor of "input_fn",
tensorflower-gardener Mar 3, 2017
02762f0
Mark Exporter and SessionBundle as deprecated. Add recommendation to …
sukritiramesh Mar 3, 2017
1708b92
Merge pull request #8066 from jhseu/branch_149155199
jhseu Mar 3, 2017
c34c4b8
Set the default flag for using nonfused winograd convolution in fowra…
tensorflower-gardener Mar 3, 2017
a0802b5
Add the graphdef version to InferenceContext and to ShapeRefiner::Add…
tensorflower-gardener Mar 3, 2017
95bb18c
Disable flaky pandas_io_test
gunan Mar 4, 2017
26f976d
Merge pull request #8065 from jhseu/r1.0_cherrypicks
Mar 4, 2017
9c86d93
Android: move makefile libtensorflow_inference.so build to sub_makefi…
andrewharp Mar 4, 2017
15c5bb2
Running the update_version.sh script to update all version #s to 1.0.1.
Mar 4, 2017
6cffd13
Merge branch 'r1.0' into cwhipkey/reduction_shapes
cwhipkey Mar 4, 2017
ef7d75b
Fix code that ignores tensorflow::Status.
Mar 4, 2017
5e0c2f3
Merge pull request #8070 from tensorflow/cwhipkey/reduction_shapes
cwhipkey Mar 4, 2017
cdf2522
Merge branch 'r1.0' into 1.0.1_patch
jhseu Mar 4, 2017
2990e81
Merge pull request #8069 from av8ramit/1.0.1_patch
jhseu Mar 4, 2017
8ed622e
change learn.metric_spec.MetricSpec to learn.MetricSpec due to API ch…
Mar 4, 2017
ba787bf
update doc link in the error msg upon import failure
rasbt Mar 4, 2017
d3f2e6c
change tf.Supervisor to tf.train.Supervisor after api changed
fanlu Mar 4, 2017
e69c83d
Fixed build error.
conqerAtapple Mar 4, 2017
1f8eae9
Fix broken link in code example, pointed out by:
tensorflower-gardener Mar 4, 2017
94634a8
Update retrain.py
chorig9 Mar 4, 2017
01c7a25
Reduce windows build logging.
tensorflower-gardener Mar 4, 2017
c19dda0
Fix some typos in the documation (tensorflow/docs_src)
yongtang Mar 4, 2017
955f038
Remove extra exported symbols.
Mar 4, 2017
2d4a15f
Merge commit for internal changes
Mar 4, 2017
f9a2c1d
Merge pull request #8086 from yongtang/03042017-docs-typo
caisq Mar 5, 2017
7e9f2dc
Merge pull request #8088 from drpngx/branch_149211889
drpngx Mar 5, 2017
d483997
Manually rollback buffer forwarding for tf.add_n().
mrry Mar 5, 2017
55238b7
remove extraneous linebreaks
rasbt Mar 5, 2017
c565660
Merge pull request #8083 from chorig9/master
caisq Mar 5, 2017
11099a0
Do not build nccl on windows. (#8048)
gunan Mar 5, 2017
01afda3
Merge pull request #8072 from wangg12/master
caisq Mar 5, 2017
87a6724
Merge pull request #8038 from cancan101/accuracy_name
caisq Mar 5, 2017
2289063
change url to https://www.tensorflow.org/install/
rasbt Mar 5, 2017
96d555e
Remove extra exported symbols.
Mar 4, 2017
e8b9bd9
Install numpy from source for Python 3.5 to fix the nightly build.
Feb 13, 2017
74ab5f1
Bump the installed protobuf for the python 3.5 build to protobuf 3.1.0
Mar 5, 2017
ac639cf
Fix link in documentation
senden9 Mar 5, 2017
35489ec
Merge pull request #8007 from qjivy/qjivy-pr
caisq Mar 5, 2017
515511d
Merge pull request #8077 from fanlu/master
caisq Mar 5, 2017
5b4bb03
Merge pull request #8097 from jhseu/r1.0_fixes
jhseu Mar 5, 2017
ff9682b
Fix broken links in documentation adding_an_op.md `g3doc/how_tos` -> …
yongtang Mar 5, 2017
3fbfdd5
Fix a bug of initializing
Jiaming-Liu Mar 5, 2017
fd46c33
Merge pull request #8028 from cberner/fix_num_servers
caisq Mar 6, 2017
2ea1fbb
Merge pull request #8081 from conqer/fix_build
caisq Mar 6, 2017
a1584b1
fix a typo in tf.conf docstring
yjmade Mar 6, 2017
6bc5d92
Internal change.
tensorflower-gardener Mar 6, 2017
5416faa
Embedding Projector, minor fixes:
Mar 6, 2017
e6f547e
Merge pull request #8128 from yjmade/patch-3
caisq Mar 6, 2017
540e8d8
Fix link in import error message.
martinwicke Mar 6, 2017
5d2a7d9
Remove the extra linkopt in ci_parameterized_build script.
gunan Mar 6, 2017
988a443
Expose bijectors as part of the distributions API
ebrevdo Mar 6, 2017
5f1976d
Android: Update gradle plugin to 2.3.0, change build.gradle API level…
andrewharp Mar 6, 2017
4763edf
Merge pull request #8135 from gunan/cp
Mar 6, 2017
19dd934
[TF:XLA] Add support for 3D convolution to XLA bridge.
hawkinsp Mar 6, 2017
8746f8a
Merge pull request #8075 from rasbt/failimport_msg
caisq Mar 6, 2017
08bc103
Re-enable buffer forwarding for GPU memory. After consulting with NVI…
tensorflower-gardener Mar 6, 2017
1ba1aaf
Update protobuf commit
Mar 6, 2017
5254f59
Add NoGradients for QuantizeAndDequantizeV2Op's input_min and input_m…
Mar 6, 2017
1f9518b
[XLA] Not clobber files when dumping HLO dot graph
tensorflower-gardener Mar 6, 2017
d41077c
* Fixed multiple formatting bugs.
tensorflower-gardener Mar 6, 2017
9ccda39
handle in SymbolicGradientOp shouldn't be accessible from multiple th…
alextp Mar 6, 2017
1d8934f
Add convenience methods to ShapeInference: MakeShapeFromTensorShape()…
hawkinsp Mar 6, 2017
96c0d2c
Fixes the treatment of collections=[] to be consistent with the docum…
tensorflower-gardener Mar 7, 2017
b397419
Automated rollback of change 149337075
gunan Mar 7, 2017
6bac753
Android: Makefile build functionality added to build.gradle to fully …
andrewharp Mar 7, 2017
020501c
Hoist resource variable reads outside functions.
alextp Mar 7, 2017
c25a662
Record memory deallocation for destroying temp variable op.
Mar 7, 2017
eaffcab
Update install instruction links. Fixes #8142.
tensorflower-gardener Mar 7, 2017
1bfe682
Add a system for TensorBoard plugin asset management to TensorFlow.
Mar 7, 2017
c177593
Add Constant, Rng, and scalars to be moved along with reshapes in res…
blakehechtman Mar 7, 2017
e17f4a6
Add an approximate_equal op for floating point tensor comparison.
Mar 7, 2017
c7b947d
Update variable_scope.py
Jiaming-Liu Mar 7, 2017
f850f48
Update ops-related pbtxt files.
tensorflower-gardener Mar 7, 2017
78d658e
Merge pull request #8104 from yongtang/8044-g3doc-how_tos-examples
caisq Mar 7, 2017
4e0c3f3
Go: Update generated wrapper functions for TensorFlow ops.
tensorflower-gardener Mar 7, 2017
6480905
[XLA] Add CallGraph::VisitNodes for traversing the call graph.
meheffernan Mar 7, 2017
3d72534
Allows users to provide custom loss function for multclass and multil…
Mar 7, 2017
6488dd4
Update variable_scope.py
Jiaming-Liu Mar 7, 2017
a61ee76
fix tf.Saver to tr.train.Saver after api changed
fanlu Mar 7, 2017
b41aae2
Fix typo in map_fn docs
guoguo12 Mar 7, 2017
2823db4
Make queue runner accept run arguments.
Mar 7, 2017
ab9ab89
[Windows/CMake] Enable tf.contrib.factorization.
mrry Mar 7, 2017
4be70b9
Go: Handle nil values for list-valued attributes gracefully.
asimshankar Mar 7, 2017
3f8a6c2
Add support for the matmul (@) operator. (#8032)
ageron Mar 7, 2017
6a3d4ab
Re-enable buffer forwarding for add_n. Add a unit test for add_n.
tensorflower-gardener Mar 7, 2017
ceb7fc1
Fix timeout for matmul_op_test in Windows.
tensorflower-gardener Mar 7, 2017
738143e
[TF:XLA] Remove support for client-allocated result buffers.
hawkinsp Mar 7, 2017
bb5cae6
Add note to new op and source build doc (#6473) (#6997)
MircoT Mar 7, 2017
df76066
Merge pull request #8157 from guoguo12/functional-ops-typo
caisq Mar 7, 2017
4c3bb1a
Merge pull request #8155 from fanlu/master
caisq Mar 7, 2017
ceff4d4
Re-enable pandas_io_test in core
Mar 7, 2017
150a0ae
Migrate quantiles library to tensorflow/contrib/boosted_trees.
tensorflower-gardener Mar 7, 2017
21bdcf5
Removed references to dynamic_rnn_estimator.RNNKeys.
johntbates Mar 7, 2017
59b3144
Link in high performance dataflow ops to grpc_tensorflow_server
saeta Mar 7, 2017
9da72b2
Merge pull request #8163 from asimshankar/go-attrs
caisq Mar 7, 2017
b7adae6
Add description how to generate ops wrapper (#7626)
Lewuathe Mar 7, 2017
b59b904
Create non-crashy versions of Concat() and Split(), to use in serving…
Mar 7, 2017
a727ee3
Closes #8165 -- Removing traces of os_setup.md (#8169)
MicaelCarvalho Mar 7, 2017
68e5a07
Use tensordot for broadcasting in Dense#call (#7650)
Mar 7, 2017
2acba51
[Java] Ability to load from SavedModel (#8148)
asimshankar Mar 7, 2017
258d0a9
Fix confusing backwards error message in tf.cond.
fwiffo Mar 7, 2017
e3fd445
Merge commit for internal changes
Mar 7, 2017
78756dc
Expose node_def from InferenceContext.
tensorflower-gardener Mar 7, 2017
19fa47b
Cleanup iOS Camera Example (#8015)
jgavris Mar 7, 2017
c82894c
Fix Linux pip install paths and instructions (#8067)
guoguo12 Mar 7, 2017
778de47
Use vector constructor to initialize Array2D. (#8115)
conqerAtapple Mar 7, 2017
671349e
Merge branch 'master' into branch_149461652
teamdandelion Mar 7, 2017
6e19f9f
add function get_collection for scope (#7977)
suiyuan2009 Mar 7, 2017
2f3251e
Expose param_regularizers in python batch_norm layer (#7390)
cancan101 Mar 7, 2017
1239d57
Update iris.py (#7115)
hprovenza Mar 7, 2017
b69dd29
EstimatorSpec: Define default scaffold in case of None provided. This…
ispirmustafa Mar 8, 2017
0b14ae4
[XLA] Remove LocalClient::ExecuteLocally(), in lieu of LocalClient::C…
hawkinsp Mar 8, 2017
53e3c4f
Merge pull request #8179 from dandelionmane/branch_149461652
teamdandelion Mar 8, 2017
95df7a3
Fix the bug that zero-sized tensor usually doesn't have an alloc id.
Mar 8, 2017
8633b22
TF-XLA Bridge ops, tests, and registrations for ReverseOp and Reverse…
tensorflower-gardener Mar 8, 2017
fbcb58b
Go: Fix #7175
asimshankar Mar 8, 2017
19fa0b8
Go: Update generated wrapper functions for TensorFlow ops.
tensorflower-gardener Mar 8, 2017
e05e5ed
Use vector constructor to initialize Array3D. (#8144)
conqerAtapple Mar 8, 2017
828a41f
Update protobuf to commit 2b7430d
Mar 8, 2017
a3e9750
Add more debug information to pandas_io_test
Mar 8, 2017
ae1668f
Correct shape->dense_shape in SparseTensor
jasonmorton Mar 8, 2017
6f00d6a
Add missing unit test case for tensordot and fix typos in implementat…
tensorflower-gardener Mar 8, 2017
2e8cf80
Merge pull request #8183 from jasonmorton/patch-1
caisq Mar 8, 2017
e83a041
Depend on protobuf's header only library when building custom ops
Mar 8, 2017
b03a72c
Add tf.spectral, a module for spectral operations.
rryan Mar 8, 2017
daaa7e5
Go: Update generated wrapper functions for TensorFlow ops.
tensorflower-gardener Mar 8, 2017
9781be4
Update ops-related pbtxt files.
tensorflower-gardener Mar 8, 2017
d357303
Make health pill text round ints.
tensorflower-gardener Mar 8, 2017
c5238fb
[TF:XLA] Implement 3d pooling ops in the TF/XLA bridge.
hawkinsp Mar 8, 2017
52c2c6b
Fix compile warnings.
conqerAtapple Mar 8, 2017
c1517aa
Adding support for int64->bool mapping in MutableDenseHashTable
rohan100jain Mar 8, 2017
b599278
[TF:XLA:GPU] Avoid a CPU-GPU synchronization when running XLA GPU com…
hawkinsp Mar 8, 2017
b2a4a7d
Java: Avoid some compiler and deprecation warnings
asimshankar Mar 8, 2017
96cb8f8
Started to open source Grappler. First application is the GPU layout …
benoitsteiner Mar 8, 2017
4ce3574
Fixing built jar location in readme
nirajpatel Mar 8, 2017
ab15148
Update installation documentation for v1.0.1.
gunan Mar 8, 2017
7138ada
Update gmock from 1.7.0 to 1.8.0
Mar 8, 2017
ce016c8
Add initial support for NCHW format for DepthwiseConv2D (and separabl…
Mar 8, 2017
34c3835
Update ops-related pbtxt files.
tensorflower-gardener Mar 8, 2017
5ac7317
Go: Update generated wrapper functions for TensorFlow ops.
tensorflower-gardener Mar 8, 2017
169a096
Merge commit for internal changes
Mar 8, 2017
d53ac1c
Make configure not create -e files on Mac (#8215)
jart Mar 8, 2017
99e30bc
Merge pull request #8212 from gunan/docs_fix
teamdandelion Mar 8, 2017
83cd3fd
Merge pull request #8106 from Jiaming-Liu/patch-1
teamdandelion Mar 8, 2017
dbd7db2
Merge pull request #8099 from senden9/patch-1
teamdandelion Mar 8, 2017
79b6914
gmock update: fix tab issue, add mirror
Mar 8, 2017
221c28d
CMake: use copy_if_different to replace copy
snnn Mar 8, 2017
2d23960
Merge pull request #8216 from dandelionmane/branch_149567254
teamdandelion Mar 8, 2017
bfb9ab4
Merging r1.0 commits back into master.
Mar 9, 2017
096e963
Fixing the Windows links in the README.
Mar 9, 2017
e10c0b4
Add cmake build for benchmark_model
snnn Mar 8, 2017
d3b5a99
Reverting this file back for now.
Mar 9, 2017
78a696b
Reverting the change to install_deb_packages too.
Mar 9, 2017
100552f
Merge pull request #8206 from conqer/fix_xla_compiler_warnings
caisq Mar 9, 2017
9f18278
Added missing SVM in init file
terrytangyuan Mar 9, 2017
d4c411e
Fix documentation in dynamic_bidirectional_rnn
erfannoury Mar 9, 2017
c25ecb5
Merge pull request #8230 from tensorflow/terrytangyuan-patch-1
caisq Mar 9, 2017
c8183de
Merge pull request #8210 from nirajpatel/patch-1
caisq Mar 9, 2017
575f6d0
Merge pull request #8237 from erfannoury/patch-1
caisq Mar 9, 2017
723288c
Merge pull request #8223 from av8ramit/r1.0_to_master_2
gunan Mar 9, 2017
eccd9fc
Merge pull request #8013 from tomergafner/fixes_ios_camera
teamdandelion Mar 9, 2017
f7ed068
Merge pull request #8213 from bogatyy/master
teamdandelion Mar 9, 2017
43c2d68
Expose SKCompat. (#8254)
martinwicke Mar 9, 2017
adc0929
Windows loadlibrary (#8217)
guschmue Mar 10, 2017
f5c700f
Merge pull request #8200 from snnn/benchmark_model
caisq Mar 10, 2017
c472fe3
sync with upstream
MycChiu Mar 10, 2017
08aa3e3
initial commit, partially finished dynamic_stich kernel
MycChiu Feb 22, 2017
5595f7b
initial commit, partially finished dynamic_stich kernel
MycChiu Feb 22, 2017
dee5112
Merge branch 'MycChiu-dynamic_stitch_GPU-patch-1' of https://github.c…
MycChiu Mar 10, 2017
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
28 changes: 0 additions & 28 deletions BUILD
@@ -1,28 +0,0 @@
# Description:
# TensorFlow is an open source software library for numerical computation using
# data flow graphs.

package(
default_visibility = [
"//tensorflow:internal",
"//tensorflow_models:__subpackages__",
],
)

licenses(["notice"]) # Apache 2.0

exports_files(["LICENSE"])

load(
"//tensorflow:tensorflow.bzl",
"cc_header_only_library",
)

cc_header_only_library(
name = "protobuf_headers",
includes = ["external/protobuf/src"],
visibility = ["//visibility:public"],
deps = [
"@protobuf//:protobuf",
],
)
14 changes: 8 additions & 6 deletions README.md
Expand Up @@ -25,18 +25,20 @@ guidelines](CONTRIBUTING.md).**

**We use [GitHub issues](https://github.com/tensorflow/tensorflow/issues) for
tracking requests and bugs, but please see
[Community](tensorflow/g3doc/resources/index.md#community) for general questions
[Community](tensorflow/docs_src/about/index.md#community) for general questions
and discussion.**

## Installation
*See [Download and Setup](tensorflow/g3doc/get_started/os_setup.md) for instructions on how to install our release binaries or how to build from source.*
*See [Installing TensorFlow](https://www.tensorflow.org/install/) for instructions on how to install our release binaries or how to build from source.*

People who are a little more adventurous can also try our nightly binaries:

* Linux CPU-only: [Python 2](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=cpu-slave/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.0-cp27-none-linux_x86_64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=cpu-slave)) / [Python 3.5](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=cpu-slave/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.0-cp35-cp35m-linux_x86_64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=cpu-slave))
* Linux GPU: [Python 2](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-linux-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=gpu-linux/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow_gpu-1.0.0-cp27-none-linux_x86_64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-linux-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=gpu-linux/)) / [Python 3.5](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-linux-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=gpu-linux/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow_gpu-1.0.0-cp35-cp35m-linux_x86_64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-linux-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=gpu-linux/))
* Mac CPU-only: [Python 2](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=mac-slave/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.0-py2-none-any.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=mac-slave/)) / [Python 3](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=mac-slave/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.0-py3-none-any.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=mac-slave/))
* Mac GPU: [Python 2](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-mac-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=gpu-mac/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow_gpu-1.0.0-py2-none-any.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-mac-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=gpu-mac/)) / [Python 3](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-mac-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=gpu-mac/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow_gpu-1.0.0-py3-none-any.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-mac-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=gpu-mac/))
* Linux CPU-only: [Python 2](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=cpu-slave/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.1-cp27-none-linux_x86_64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=cpu-slave)) / [Python 3.4](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=cpu-slave/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.1-cp34-cp34m-linux_x86_64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=cpu-slave/)) / [Python 3.5](https://ci.tensorflow.org/view/Nightly/job/nightly-python35-linux-cpu/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.1-cp35-cp35m-linux_x86_64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-python35-linux-cpu/))
* Linux GPU: [Python 2](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-linux-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=gpu-linux/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow_gpu-1.0.1-cp27-none-linux_x86_64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-linux-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=gpu-linux/)) / [Python 3.4](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-linux-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=gpu-linux/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow_gpu-1.0.1-cp34-cp34m-linux_x86_64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-linux-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=gpu-linux/)) / [Python 3.5](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-linux-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3.5,label=gpu-linux/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow_gpu-1.0.1-cp35-cp35m-linux_x86_64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-linux-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3.5,label=gpu-linux/))
* Mac CPU-only: [Python 2](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=mac-slave/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.1-py2-none-any.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=mac-slave/)) / [Python 3](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=mac-slave/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow-1.0.1-py3-none-any.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=mac-slave/))
* Mac GPU: [Python 2](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-mac-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=gpu-mac/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow_gpu-1.0.1-py2-none-any.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-mac-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON2,label=gpu-mac/)) / [Python 3](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-mac-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=gpu-mac/lastSuccessfulBuild/artifact/pip_test/whl/tensorflow_gpu-1.0.1-py3-none-any.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-matrix-mac-gpu/TF_BUILD_IS_OPT=OPT,TF_BUILD_IS_PIP=PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=gpu-mac/))
* Windows CPU-only: [Python 3.5 64-bit](https://ci.tensorflow.org/view/Nightly/job/nightly-win/DEVICE=cpu,OS=windows/lastSuccessfulBuild/artifact/cmake_build/tf_python/dist/tensorflow-1.0.1-cp35-cp35m-win_amd64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-win/DEVICE=cpu,OS=windows/))
* Windows GPU: [Python 3.5 64-bit](https://ci.tensorflow.org/view/Nightly/job/nightly-win/DEVICE=gpu,OS=windows/lastSuccessfulBuild/artifact/cmake_build/tf_python/dist/tensorflow_gpu-1.0.1-cp35-cp35m-win_amd64.whl) ([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-win/DEVICE=gpu,OS=windows/))
* Android: [demo APK](https://ci.tensorflow.org/view/Nightly/job/nightly-android/lastSuccessfulBuild/artifact/out/tensorflow_demo.apk), [native libs](http://ci.tensorflow.org/view/Nightly/job/nightly-android/lastSuccessfulBuild/artifact/out/native/)
([build history](https://ci.tensorflow.org/view/Nightly/job/nightly-android/))

Expand Down
1 change: 1 addition & 0 deletions RELEASE.md
Expand Up @@ -87,6 +87,7 @@ To help you upgrade your existing TensorFlow Python code to match the API change
* In the C++ API (in tensorflow/cc), Input, Output, etc. have moved
from the tensorflow::ops namespace to tensorflow.
* Change arg order for `{softmax,sparse_softmax,sigmoid}_cross_entropy_with_logits` to be (labels, predictions), and force use of named args.
* tf.nn.rnn_cell.* and most functions in tf.nn.rnn.* (with the exception of dynamic_rnn and raw_rnn) are temporarily in tf.contrib.rnn. They will be moved back into core for TF 1.1.

## Bug Fixes and Other Changes
* Numerous C++ API updates.
Expand Down
4 changes: 2 additions & 2 deletions WORKSPACE
Expand Up @@ -25,15 +25,15 @@ check_version("0.4.2")
#android_sdk_repository(
# name = "androidsdk",
# api_level = 23,
# build_tools_version = "23.0.1",
# build_tools_version = "25.0.1",
# # Replace with path to Android SDK on your system
# path = "<PATH_TO_SDK>",
#)
#
#android_ndk_repository(
# name="androidndk",
# path="<PATH_TO_NDK>",
# api_level=21)
# api_level=14)

# Please add all new TensorFlow dependencies in workspace.bzl.
tf_workspace()
Expand Down
24 changes: 16 additions & 8 deletions configure
Expand Up @@ -45,6 +45,14 @@ function bazel_clean_and_fetch() {
-//tensorflow/examples/android/..."
}

function sed_hyphen_i() {
if is_macos; then
sed -i '' "$@"
else
sed -i "$@"
fi
}

# Delete any leftover BUILD files from the Makefile build, which would interfere
# with Bazel parsing.
MAKEFILE_DOWNLOAD_DIR=tensorflow/contrib/makefile/downloads
Expand Down Expand Up @@ -173,9 +181,9 @@ else
fi

if [ "$TF_NEED_JEMALLOC" == "1" ]; then
sed -i -e "s/WITH_JEMALLOC = False/WITH_JEMALLOC = True/" tensorflow/core/platform/default/build_config.bzl
sed_hyphen_i -e "s/WITH_JEMALLOC = False/WITH_JEMALLOC = True/" tensorflow/core/platform/default/build_config.bzl
else
sed -i -e "s/WITH_JEMALLOC = True/WITH_JEMALLOC = False/" tensorflow/core/platform/default/build_config.bzl
sed_hyphen_i -e "s/WITH_JEMALLOC = True/WITH_JEMALLOC = False/" tensorflow/core/platform/default/build_config.bzl
fi

while [ "$TF_NEED_GCP" == "" ]; do
Expand All @@ -202,10 +210,10 @@ if [ "$TF_NEED_GCP" == "1" ]; then
fi

# Update Bazel build configuration.
sed -i -e "s/WITH_GCP_SUPPORT = False/WITH_GCP_SUPPORT = True/" tensorflow/core/platform/default/build_config.bzl
sed_hyphen_i -e "s/WITH_GCP_SUPPORT = False/WITH_GCP_SUPPORT = True/" tensorflow/core/platform/default/build_config.bzl
else
# Update Bazel build configuration.
sed -i -e "s/WITH_GCP_SUPPORT = True/WITH_GCP_SUPPORT = False/" tensorflow/core/platform/default/build_config.bzl
sed_hyphen_i -e "s/WITH_GCP_SUPPORT = True/WITH_GCP_SUPPORT = False/" tensorflow/core/platform/default/build_config.bzl
fi

while [ "$TF_NEED_HDFS" == "" ]; do
Expand All @@ -224,10 +232,10 @@ done

if [ "$TF_NEED_HDFS" == "1" ]; then
# Update Bazel build configuration.
sed -i -e "s/WITH_HDFS_SUPPORT = False/WITH_HDFS_SUPPORT = True/" tensorflow/core/platform/default/build_config.bzl
sed_hyphen_i -e "s/WITH_HDFS_SUPPORT = False/WITH_HDFS_SUPPORT = True/" tensorflow/core/platform/default/build_config.bzl
else
# Update Bazel build configuration.
sed -i -e "s/WITH_HDFS_SUPPORT = True/WITH_HDFS_SUPPORT = False/" tensorflow/core/platform/default/build_config.bzl
sed_hyphen_i -e "s/WITH_HDFS_SUPPORT = True/WITH_HDFS_SUPPORT = False/" tensorflow/core/platform/default/build_config.bzl
fi

## Enable XLA.
Expand All @@ -243,10 +251,10 @@ done

if [ "$TF_ENABLE_XLA" == "1" ]; then
# Update Bazel build configuration.
sed -i -e "s/^WITH_XLA_SUPPORT = [FT].*/WITH_XLA_SUPPORT = True/" tensorflow/core/platform/default/build_config_root.bzl
sed_hyphen_i -e "s/^WITH_XLA_SUPPORT = [FT].*/WITH_XLA_SUPPORT = True/" tensorflow/core/platform/default/build_config_root.bzl
else
# Update Bazel build configuration.
sed -i -e "s/^WITH_XLA_SUPPORT = [FT].*/WITH_XLA_SUPPORT = False/" tensorflow/core/platform/default/build_config_root.bzl
sed_hyphen_i -e "s/^WITH_XLA_SUPPORT = [FT].*/WITH_XLA_SUPPORT = False/" tensorflow/core/platform/default/build_config_root.bzl
fi


Expand Down
17 changes: 8 additions & 9 deletions tensorflow/BUILD
Expand Up @@ -147,7 +147,9 @@ filegroup(
"//tensorflow/compiler/aot/tests:all_files",
"//tensorflow/compiler/jit:all_files",
"//tensorflow/compiler/jit/graphcycles:all_files",
"//tensorflow/compiler/jit/kernels:all_files",
"//tensorflow/compiler/jit/legacy_flags:all_files",
"//tensorflow/compiler/jit/ops:all_files",
"//tensorflow/compiler/tests:all_files",
"//tensorflow/compiler/tf2xla:all_files",
"//tensorflow/compiler/tf2xla/kernels:all_files",
Expand Down Expand Up @@ -219,6 +221,11 @@ filegroup(
"//tensorflow/core/debug:all_files",
"//tensorflow/core/distributed_runtime:all_files",
"//tensorflow/core/distributed_runtime/rpc:all_files",
"//tensorflow/core/grappler:all_files",
"//tensorflow/core/grappler/clusters:all_files",
"//tensorflow/core/grappler/costs:all_files",
"//tensorflow/core/grappler/inputs:all_files",
"//tensorflow/core/grappler/optimizers:all_files",
"//tensorflow/core/kernels:all_files",
"//tensorflow/core/kernels/cloud:all_files",
"//tensorflow/core/kernels/hexagon:all_files",
Expand All @@ -243,6 +250,7 @@ filegroup(
"//tensorflow/java/src/main/native:all_files",
"//tensorflow/python:all_files",
"//tensorflow/python/debug:all_files",
"//tensorflow/python/estimator:all_files",
"//tensorflow/python/kernel_tests:all_files",
"//tensorflow/python/saved_model:all_files",
"//tensorflow/python/tools:all_files",
Expand Down Expand Up @@ -305,15 +313,6 @@ cc_binary(
],
)

cc_binary(
name = "libtensorflow_c.so",
linkshared = 1,
deps = [
"//tensorflow/c:c_api",
"//tensorflow/core:tensorflow",
],
)

cc_binary(
name = "libtensorflow_cc.so",
linkshared = 1,
Expand Down
13 changes: 13 additions & 0 deletions tensorflow/__init__.py
Expand Up @@ -37,3 +37,16 @@ def __getattr__(self, item):


contrib = _LazyContribLoader()

del absolute_import
del division
del print_function

# These symbols appear because we import the python package which
# in turn imports from tensorflow.core and tensorflow.python. They
# must come from this module. So python adds these symbols for the
# resolution to succeed.
# pylint: disable=undefined-variable
del python
del core
# pylint: enable=undefined-variable
30 changes: 21 additions & 9 deletions tensorflow/c/c_api.cc
Expand Up @@ -543,9 +543,8 @@ static void TF_Run_Helper(

if (handle == nullptr) {
RunOptions run_options_proto;
if (run_options != nullptr &&
!run_options_proto.ParseFromArray(run_options->data,
run_options->length)) {
if (run_options != nullptr && !run_options_proto.ParseFromArray(
run_options->data, run_options->length)) {
status->status = InvalidArgument("Unparseable RunOptions proto");
return;
}
Expand Down Expand Up @@ -731,7 +730,7 @@ extern "C" {
struct TF_Graph {
TF_Graph()
: graph(OpRegistry::Global()),
refiner(graph.op_registry()),
refiner(graph.versions().producer(), graph.op_registry()),
num_sessions(0),
delete_requested(false),
parent(nullptr),
Expand Down Expand Up @@ -1702,6 +1701,12 @@ void TF_ImportGraphDefOptionsAddInputMapping(TF_ImportGraphDefOptions* opts,
opts->opts.input_map[TensorId(src_name, src_index)] = ToTensorId(dst);
}

void TF_ImportGraphDefOptionsRemapControlDependency(
TF_ImportGraphDefOptions* opts, const char* src_name, TF_Operation* dst) {
opts->opts.input_map[TensorId(src_name, tensorflow::Graph::kControlSlot)] =
TensorId(dst->node.name(), tensorflow::Graph::kControlSlot);
}

extern void TF_ImportGraphDefOptionsAddControlDependency(
TF_ImportGraphDefOptions* opts, TF_Operation* oper) {
opts->opts.control_dependencies.push_back(oper->node.name());
Expand Down Expand Up @@ -2111,11 +2116,19 @@ TF_Session* TF_NewSession(TF_Graph* graph, const TF_SessionOptions* opt,
}
}

#ifndef __ANDROID__
TF_Session* TF_LoadSessionFromSavedModel(
const TF_SessionOptions* session_options, const TF_Buffer* run_options,
const char* export_dir, const char* const* tags, int tags_len,
TF_Graph* graph, TF_Buffer* meta_graph_def, TF_Status* status) {
// TODO(ashankar): Remove the __ANDROID__ guard. This will require ensuring that
// the tensorflow/cc/saved_model:loader build target is Android friendly.
#ifdef __ANDROID__
status->status = tensorflow::errors::Unimplemented(
"Loading a SavedModel is not supported in Android. File a bug at "
"https://github.com/tensorflow/tensorflow/issues if this feature is "
"important to you");
return nullptr;
#else
mutex_lock l(graph->mu);

if (!graph->name_map.empty()) {
Expand All @@ -2124,9 +2137,8 @@ TF_Session* TF_LoadSessionFromSavedModel(
}

RunOptions run_options_proto;
if (run_options != nullptr &&
!run_options_proto.ParseFromArray(run_options->data,
run_options->length)) {
if (run_options != nullptr && !run_options_proto.ParseFromArray(
run_options->data, run_options->length)) {
status->status = InvalidArgument("Unparseable RunOptions proto");
return nullptr;
}
Expand Down Expand Up @@ -2164,8 +2176,8 @@ TF_Session* TF_LoadSessionFromSavedModel(
graph->num_sessions += 1;
session->last_num_graph_nodes = graph->graph.num_node_ids();
return session;
}
#endif // __ANDROID__
}

void TF_CloseSession(TF_Session* s, TF_Status* status) {
status->status = s->session->Close();
Expand Down