Mark Daoust [Thu, 1 Feb 2018 18:43:29 +0000 (10:43 -0800)]
Add function paths to their signatures.
fixes #16167
PiperOrigin-RevId:
184160925
A. Unique TensorFlower [Thu, 1 Feb 2018 18:38:14 +0000 (10:38 -0800)]
Fix nest bug with different dictionary key orderings.
PiperOrigin-RevId:
184160009
A. Unique TensorFlower [Thu, 1 Feb 2018 17:55:53 +0000 (09:55 -0800)]
Add shape inference for outside_compilation graph rewrite. Pull out enough of the graph to enable inference of the shape of a SendFromHost Op once the shape of corresponding RecvAtHost Ops are known.
PiperOrigin-RevId:
184153187
A. Unique TensorFlower [Thu, 1 Feb 2018 16:18:13 +0000 (08:18 -0800)]
Update ops-related pbtxt files.
PiperOrigin-RevId:
184141875
Peter Hawkins [Thu, 1 Feb 2018 14:47:06 +0000 (06:47 -0800)]
[TF:XLA] Implement MatrixSetDiag and MatrixBandPart.
Add support for int32 indices to the MatrixBandPart operator.
PiperOrigin-RevId:
184133343
Sanjoy Das [Thu, 1 Feb 2018 07:05:26 +0000 (23:05 -0800)]
[TF:XLA] Fix tfcompile OSS build
- The @org_tensorflow package designation is unnecessary, and breaks the build
when building without a sandbox.
- The generated tests must use tf_cc_test, not cc_test. See the note in
tensorflow/core/BUILD.
Partially addresses #15338
PiperOrigin-RevId:
184095571
Yu-Cheng Ling [Thu, 1 Feb 2018 05:17:30 +0000 (21:17 -0800)]
Internal change
PiperOrigin-RevId:
184088913
A. Unique TensorFlower [Thu, 1 Feb 2018 04:46:33 +0000 (20:46 -0800)]
Go: Update generated wrapper functions for TensorFlow ops.
PiperOrigin-RevId:
184086955
A. Unique TensorFlower [Thu, 1 Feb 2018 04:20:18 +0000 (20:20 -0800)]
Update ops-related pbtxt files.
PiperOrigin-RevId:
184085402
Eugene Brevdo [Thu, 1 Feb 2018 02:18:04 +0000 (18:18 -0800)]
Add a new Dataset: PrependFromQueueAndPaddedBatchDataset.
PiperOrigin-RevId:
184078894
Yu-Cheng Ling [Thu, 1 Feb 2018 02:08:33 +0000 (18:08 -0800)]
Test all TFLite kernel implementations (reference/optimized/...)
PiperOrigin-RevId:
184077940
Akshay Agrawal [Thu, 1 Feb 2018 02:03:25 +0000 (18:03 -0800)]
TFE: Support `IndexedSlices` as inputs and outputs for `tfe.defun`.
In particular, this change fixes a bug that arose when attempting to
differentiate a defun-d function whose gradient yielded `IndexedSlices`.
PiperOrigin-RevId:
184077261
Mark Daoust [Thu, 1 Feb 2018 01:46:43 +0000 (17:46 -0800)]
Add a doc explaining how to convert an `Estimator` to run on a Cloud TPU.
PiperOrigin-RevId:
184075365
Roy Frostig [Thu, 1 Feb 2018 00:54:20 +0000 (16:54 -0800)]
[XLA] Support array layout specification in local Python XLA client.
PiperOrigin-RevId:
184069454
Derek Murray [Thu, 1 Feb 2018 00:40:10 +0000 (16:40 -0800)]
Add type check when assigning to a resource variable slice.
PiperOrigin-RevId:
184067663
Asim Shankar [Thu, 1 Feb 2018 00:17:42 +0000 (16:17 -0800)]
eager: Fix dropout in MNIST example.
The dropout layer takes a training argument in __call__, which defaults
to false. So without this change, dropout was not being applied.
PiperOrigin-RevId:
184064465
Tatiana Shpeisman [Wed, 31 Jan 2018 23:47:42 +0000 (15:47 -0800)]
MKL is no longer enabled via ./configure. Fixed documentation to reflect this.
PiperOrigin-RevId:
184060046
Tatiana Shpeisman [Wed, 31 Jan 2018 23:41:03 +0000 (15:41 -0800)]
Change recommended option for building TensorFlow with MKL from -c opt to --config=opt. -c opt triggers optimized C++ compilation. --config=opt also uses additional optimization flags as set by running ./configure.
PiperOrigin-RevId:
184059060
A. Unique TensorFlower [Wed, 31 Jan 2018 23:32:39 +0000 (15:32 -0800)]
Update external protobuf codebase version for Windows cmake build
PiperOrigin-RevId:
184057827
Gunhan Gulsoy [Wed, 31 Jan 2018 22:57:24 +0000 (14:57 -0800)]
Enable AVX in all TF windows builds.
PiperOrigin-RevId:
184052414
Shanqing Cai [Wed, 31 Jan 2018 22:55:13 +0000 (14:55 -0800)]
Add grpcio as a pip dependency of tensorflow
PiperOrigin-RevId:
184052073
Shivani Agrawal [Wed, 31 Jan 2018 22:35:37 +0000 (14:35 -0800)]
Fixes minor typos.
PiperOrigin-RevId:
184048812
Nick Desaulniers [Wed, 31 Jan 2018 22:32:21 +0000 (14:32 -0800)]
[XLA] Initialize linear indices used by NearComparator.
In unoptimized builds, we'd see miscompares with insanely large indices when
comparing literals containing NaN.
The linear indices may never be updated/initialized if the value NaN was
compared, since:
(NaN > x) == false
(NaN < x) == false
Adds a few tests that were used when debugging, but not a perfect one for this
case. It's currently not possible to test that LiteralTestUtil::Near() fails
when given bad input since it uses the EXPECT_* family of macros, which would
cause the intentional miscompares to fail the test no matter what.
PiperOrigin-RevId:
184048275
Anna R [Wed, 31 Jan 2018 22:30:22 +0000 (14:30 -0800)]
Internal change.
PiperOrigin-RevId:
184047860
A. Unique TensorFlower [Wed, 31 Jan 2018 21:42:18 +0000 (13:42 -0800)]
Tolerate Const nodes with no data or with smaller data than is required
by their shape, by zero-extending the Const data to the required size.
We wanted to generate an error on that, but too many
existing graphs already rely on current lax behavior.
PiperOrigin-RevId:
184039876
A. Unique TensorFlower [Wed, 31 Jan 2018 21:35:23 +0000 (13:35 -0800)]
Automated g4 rollback of changelist
184003263
PiperOrigin-RevId:
184038801
Sanjoy Das [Wed, 31 Jan 2018 21:22:49 +0000 (13:22 -0800)]
[XLA:CPU] Generate correct IR for integer clamp
PiperOrigin-RevId:
184037078
Peter Hawkins [Wed, 31 Jan 2018 20:58:05 +0000 (12:58 -0800)]
[TF:XLA] Implement ExtractImagePatches.
PiperOrigin-RevId:
184033616
A. Unique TensorFlower [Wed, 31 Jan 2018 20:32:38 +0000 (12:32 -0800)]
Remove contacts of ex-Googler
PiperOrigin-RevId:
184030353
Peter Hawkins [Wed, 31 Jan 2018 20:28:13 +0000 (12:28 -0800)]
[TF:XLA] Implement Acos, Asin, Atan in terms of Atan2 using half-angle formulae. This may not be the most efficient implementation but it is better than no implementation.
PiperOrigin-RevId:
184029858
Sanjoy Das [Wed, 31 Jan 2018 20:27:38 +0000 (12:27 -0800)]
[TF:XLA] Bump open source llvm revision to r323874
PiperOrigin-RevId:
184029790
Alexandre Passos [Wed, 31 Jan 2018 20:22:10 +0000 (12:22 -0800)]
Increfing tapes when keeping them around calls to python API functions.
These calls can trigger GIL releases or python GC which can trigger
a change to the set of live tapes and segmentation faults.
PiperOrigin-RevId:
184029146
Yifei Feng [Wed, 31 Jan 2018 19:34:56 +0000 (11:34 -0800)]
De-bazel filename_test. Part of the effort to remove all_opensource_files.
PiperOrigin-RevId:
184021942
Todd Wang [Wed, 31 Jan 2018 19:32:35 +0000 (11:32 -0800)]
Go: Fix Scope.WithControlDependencies array-copying behavior.
The test fails with the old code, and passes with the new code.
PiperOrigin-RevId:
184021596
Jonathan Hseu [Wed, 31 Jan 2018 19:30:36 +0000 (11:30 -0800)]
TPUEstimator: support host_call when use_tpu=False.
PiperOrigin-RevId:
184021299
Anna R [Wed, 31 Jan 2018 19:25:38 +0000 (11:25 -0800)]
Adding tf_export decorators/calls to TensorFlow functions and constants.
PiperOrigin-RevId:
184020524
A. Unique TensorFlower [Wed, 31 Jan 2018 18:59:49 +0000 (10:59 -0800)]
Typo Correction.
PiperOrigin-RevId:
184016082
Eric Liu [Wed, 31 Jan 2018 18:38:57 +0000 (10:38 -0800)]
Add std:: to min/max in cuda code to unbreak cuda-clang build.
PiperOrigin-RevId:
184012479
Shivani Agrawal [Wed, 31 Jan 2018 18:29:29 +0000 (10:29 -0800)]
Whitelisting stateful op for dataset checkpointing temporarily.
PiperOrigin-RevId:
184010966
Blake Hechtman [Wed, 31 Jan 2018 18:19:17 +0000 (10:19 -0800)]
[XLA] Disable transpose folding into reduce for reduces of rank 2 or higher.
PiperOrigin-RevId:
184009219
A. Unique TensorFlower [Wed, 31 Jan 2018 18:06:24 +0000 (10:06 -0800)]
Update external protobuf codebase version
PiperOrigin-RevId:
184006959
A. Unique TensorFlower [Wed, 31 Jan 2018 17:53:17 +0000 (09:53 -0800)]
Extending sparsify_gather to remove variables from the tensorflow summaries.
PiperOrigin-RevId:
184004859
A. Unique TensorFlower [Wed, 31 Jan 2018 17:40:48 +0000 (09:40 -0800)]
Verify contents of tensors
PiperOrigin-RevId:
184003263
Mark Daoust [Wed, 31 Jan 2018 16:16:59 +0000 (08:16 -0800)]
Fix docs generation for cluster_resolvers
Adds "cluster_resolver_pip" as a dependancy to opensource contrib, and applies a standard `remove_undocumented` to clear extra symbols.
Docs are build from a bazel bulid, and without this change the cluster resolvers are not directly accessible in "tf.contirb.cluster_resolver" during the docs build, so they do not get documented.
PiperOrigin-RevId:
183993115
Eric Liu [Wed, 31 Jan 2018 09:26:29 +0000 (01:26 -0800)]
Remove a failing test for gpu numbers when XLA is enabled.
The test assumes that all devices are either CPU or GPU, which is not true when XLA is enabled.
PiperOrigin-RevId:
183956872
Asim Shankar [Wed, 31 Jan 2018 07:15:28 +0000 (23:15 -0800)]
Go: Support control dependencies.
Fixes #16464
PiperOrigin-RevId:
183946928
Jianwei Xie [Wed, 31 Jan 2018 05:34:02 +0000 (21:34 -0800)]
Adds input_fn-return-dataset for Per-Host input pipeline deployment.
PiperOrigin-RevId:
183940666
Anna R [Wed, 31 Jan 2018 04:28:38 +0000 (20:28 -0800)]
Adding tf_export decorators/calls to TensorFlow functions and constants.
PiperOrigin-RevId:
183936100
Bjarke Hammersholt Roune [Wed, 31 Jan 2018 04:12:22 +0000 (20:12 -0800)]
Remove Google-internal bug numbers from XLA error messages.
PiperOrigin-RevId:
183934860
A. Unique TensorFlower [Wed, 31 Jan 2018 03:30:18 +0000 (19:30 -0800)]
internal change
PiperOrigin-RevId:
183932015
Anjali Sridhar [Wed, 31 Jan 2018 02:50:56 +0000 (18:50 -0800)]
Eager support for tf.Keras.
PiperOrigin-RevId:
183928850
A. Unique TensorFlower [Wed, 31 Jan 2018 02:08:52 +0000 (18:08 -0800)]
internal change
PiperOrigin-RevId:
183924520
A. Unique TensorFlower [Wed, 31 Jan 2018 02:03:31 +0000 (18:03 -0800)]
Removed unused variable doReverseChannels.
PiperOrigin-RevId:
183923876
A. Unique TensorFlower [Wed, 31 Jan 2018 01:58:28 +0000 (17:58 -0800)]
K-FAC: Wrap extract_image_patches() for compatibility with XLA.
PiperOrigin-RevId:
183923073
Max Galkin [Wed, 31 Jan 2018 01:41:00 +0000 (17:41 -0800)]
Add a CLIF wrapper for DeviceProperties.
PiperOrigin-RevId:
183921389
Jonathan Hseu [Wed, 31 Jan 2018 01:28:42 +0000 (17:28 -0800)]
Give tf.contrib.summary a default suffix of .v2. This is so that it doesn't conflict with v1 summaries that are being created at the same time.
By default, MonitoredTrainingSession and Estimator will create v1 summaries, so using both currently creates a race condition, and the last one to create the event file writer wins.
PiperOrigin-RevId:
183919988
Suharsh Sivakumar [Wed, 31 Jan 2018 01:03:49 +0000 (17:03 -0800)]
Update graph_matcher to include OneofPattern.
PiperOrigin-RevId:
183916547
Shivani Agrawal [Wed, 31 Jan 2018 00:14:17 +0000 (16:14 -0800)]
[tf.data] Native support for `tf.SparseTensor` and serialization tests for the `tf.data` methods that can produce `tf.SparseTensor` elements.
PiperOrigin-RevId:
183909820
Anjali Sridhar [Wed, 31 Jan 2018 00:12:32 +0000 (16:12 -0800)]
Fix _standardize_input_data function to allow users to use list of
ints/floats and list of lists as training and validation data input.
PiperOrigin-RevId:
183909545
A. Unique TensorFlower [Tue, 30 Jan 2018 23:53:55 +0000 (15:53 -0800)]
[XLA] Add Clamp and Round to the local Python XLA client.
PiperOrigin-RevId:
183906722
Anna R [Tue, 30 Jan 2018 23:36:08 +0000 (15:36 -0800)]
Internal change.
PiperOrigin-RevId:
183904042
Nupur Garg [Tue, 30 Jan 2018 23:11:33 +0000 (15:11 -0800)]
Internal change.
PiperOrigin-RevId:
183900332
Amit Patankar [Tue, 30 Jan 2018 23:07:03 +0000 (15:07 -0800)]
Supporting paths for build_server script.
PiperOrigin-RevId:
183899675
A. Unique TensorFlower [Tue, 30 Jan 2018 23:05:49 +0000 (15:05 -0800)]
Always evaluate() on TPU in TPUEstimator unless use_tpu is set to False.
PiperOrigin-RevId:
183899481
A. Unique TensorFlower [Tue, 30 Jan 2018 23:04:14 +0000 (15:04 -0800)]
Optimize memory allocation of TFLite StridedSlice Op for constant tensors.
PiperOrigin-RevId:
183899156
Asim Shankar [Tue, 30 Jan 2018 23:00:10 +0000 (15:00 -0800)]
Assign*VariableOp: Documentation fix.
These ops have no outputs.
The way to ensure ordering is by adding dependencies via control inputs.
Helps with #16464
PiperOrigin-RevId:
183898409
A. Unique TensorFlower [Tue, 30 Jan 2018 22:47:11 +0000 (14:47 -0800)]
Expose the decorator in the main API. Move the top-level implementation files into a submodule, to avoid problems around the remove_undocumented call.
PiperOrigin-RevId:
183896273
David Majnemer [Tue, 30 Jan 2018 22:28:06 +0000 (14:28 -0800)]
[XLA] Add a test for scalar kRoundNearestAfz
PiperOrigin-RevId:
183892483
A. Unique TensorFlower [Tue, 30 Jan 2018 22:07:26 +0000 (14:07 -0800)]
Add `tf.contrib.distributions.Kumaraswamy`.
PiperOrigin-RevId:
183888726
Todd Wang [Tue, 30 Jan 2018 22:06:07 +0000 (14:06 -0800)]
Remove unnecessary dependency from xla:types
PiperOrigin-RevId:
183888495
A. Unique TensorFlower [Tue, 30 Jan 2018 21:58:05 +0000 (13:58 -0800)]
Add and Mul support broadcasting.
PiperOrigin-RevId:
183886920
Derek Murray [Tue, 30 Jan 2018 21:26:51 +0000 (13:26 -0800)]
[tf.data] Add `IteratorContext::allocator()`.
This enables the various iterator implementations to use the actual allocator for the device on which they are running, rather than defaulting to `cpu_allocator()` (which is typically a plain malloc). In future, this will enable allocating iterator outputs in CUDA-pinned memory (and GPU memory).
PERFORMANCE NOTE: In sessions where `ConfigProto.force_gpu_compatible == True`, this change has the effect of allocating all input pipeline tensors in CUDA-pinned memory. Previous if this flag was set, only the tensors allocated during function execution would be allocated in this space, and other tensors (e.g. the result of a `Dataset.batch()` would be allocated using `cpu_allocator()` (i.e. `malloc()`). This change should lead to more efficient communication between a host-side input pipeline and GPUs, but it may also create more pressure on the CUDA host allocator (whose default maximum size is 64GB). The "TF_CUDA_HOST_MEM_LIMIT_IN_MB" environment variable can be used to override this value.
This change is a starting point for working on issue #13610.
PiperOrigin-RevId:
183881907
Asim Shankar [Tue, 30 Jan 2018 21:20:13 +0000 (13:20 -0800)]
Eager: Update documentation to reflect that you can use TensorFlow 1.5
PiperOrigin-RevId:
183880991
A. Unique TensorFlower [Tue, 30 Jan 2018 21:10:54 +0000 (13:10 -0800)]
Add missing dependency.
PiperOrigin-RevId:
183879566
Russell Power [Tue, 30 Jan 2018 21:02:25 +0000 (13:02 -0800)]
Enable bulk restoration by default.
This enables loading multiple tensors in single call, allowing for better buffering and reduced load on distributed filesystems.
PiperOrigin-RevId:
183878169
Jonathan Hseu [Tue, 30 Jan 2018 20:49:18 +0000 (12:49 -0800)]
Support outfeed host calls in TPUEstimator. That implicitly allows us to support tf.contrib.summary (see the unit test).
We may change the recommended way to use summaries later.
PiperOrigin-RevId:
183876356
A. Unique TensorFlower [Tue, 30 Jan 2018 20:36:51 +0000 (12:36 -0800)]
Automated g4 rollback of changelist
183846994
PiperOrigin-RevId:
183874527
Martin Wicke [Tue, 30 Jan 2018 20:32:30 +0000 (12:32 -0800)]
Fix bad logging call in warmstarting_utils.py.
PiperOrigin-RevId:
183873925
A. Unique TensorFlower [Tue, 30 Jan 2018 20:13:58 +0000 (12:13 -0800)]
Add check macro for not-equal.
PiperOrigin-RevId:
183871336
David G. Andersen [Tue, 30 Jan 2018 20:06:40 +0000 (12:06 -0800)]
Eliminate crash on a 'no error' return from DecodeGif when parsing an invalid
gif. (Previous code tried to strcat a null).
PiperOrigin-RevId:
183870288
A. Unique TensorFlower [Tue, 30 Jan 2018 20:01:02 +0000 (12:01 -0800)]
RetryingFileSystem::FlushCaches() calls the base FileSystem's FlushCaches().
PiperOrigin-RevId:
183869325
A. Unique TensorFlower [Tue, 30 Jan 2018 20:00:04 +0000 (12:00 -0800)]
Add static_sample flag to Mixture, permitting calls to `sample` to not rely on dynamic tensor indexing. This allows for some static graph compilation optimizations, but at the expense of sampling all underlying distributions in the mixture.
PiperOrigin-RevId:
183869189
Sanjoy Das [Tue, 30 Jan 2018 19:53:26 +0000 (11:53 -0800)]
[TF:XLA] Bump open source llvm revision to r323761
PiperOrigin-RevId:
183868087
A. Unique TensorFlower [Tue, 30 Jan 2018 19:47:43 +0000 (11:47 -0800)]
K-FAC: expose set_global_constants() for tf.contrib.kfac.utils
PiperOrigin-RevId:
183867014
Akshay Agrawal [Tue, 30 Jan 2018 19:42:31 +0000 (11:42 -0800)]
Delete dead code in Layer.
PiperOrigin-RevId:
183866106
Chris Leary [Tue, 30 Jan 2018 19:22:12 +0000 (11:22 -0800)]
[XLA] Fix tools broken by cl/
183837856
PiperOrigin-RevId:
183862522
Skye Wanderman-Milne [Tue, 30 Jan 2018 19:18:03 +0000 (11:18 -0800)]
Make loss_ops_test.py work with C API enabled.
PiperOrigin-RevId:
183861779
A. Unique TensorFlower [Tue, 30 Jan 2018 19:09:06 +0000 (11:09 -0800)]
Add py2tf to contrib_py.
PiperOrigin-RevId:
183860192
Chris Leary [Tue, 30 Jan 2018 19:22:12 +0000 (11:22 -0800)]
[XLA] Fix tools broken by cl/
183837856
PiperOrigin-RevId:
183862522
Skye Wanderman-Milne [Tue, 30 Jan 2018 19:18:03 +0000 (11:18 -0800)]
Make loss_ops_test.py work with C API enabled.
PiperOrigin-RevId:
183861779
A. Unique TensorFlower [Tue, 30 Jan 2018 19:09:06 +0000 (11:09 -0800)]
Add py2tf to contrib_py.
PiperOrigin-RevId:
183860192
A. Unique TensorFlower [Tue, 30 Jan 2018 18:43:03 +0000 (10:43 -0800)]
Cleanup: Ran clang-format on all *.{cc,h} files in tensorflow/contrib/.../*.{hh,c}.
PiperOrigin-RevId:
183855242
A. Unique TensorFlower [Tue, 30 Jan 2018 18:18:36 +0000 (10:18 -0800)]
Prepare variance to be exported for serving with the servo library.
PiperOrigin-RevId:
183851026
A. Unique TensorFlower [Tue, 30 Jan 2018 18:05:04 +0000 (10:05 -0800)]
Cleanup: Ran clang-format on files in tensorflow/core/.../*.{cc,h}.
PiperOrigin-RevId:
183848459
A. Unique TensorFlower [Tue, 30 Jan 2018 17:57:16 +0000 (09:57 -0800)]
TensorFlow SavedModel loader: avoid segmentation fault when NewSession returns null
PiperOrigin-RevId:
183846994
Andrew Selle [Tue, 30 Jan 2018 17:55:38 +0000 (09:55 -0800)]
Create an interface to create hints for future toco conversions.
Specifically, tf.contrib.lite.OpHint can create "breadcrumb"
hints that describe encapsulation of multiple TensorFlow ops
that make up a TensorFlow lite builtin or custom op. These
can later be replaced with stub versions in a GraphDef or
SavedModel.
PiperOrigin-RevId:
183846742
Yuanzhong Xu [Tue, 30 Jan 2018 17:54:46 +0000 (09:54 -0800)]
Add BF16 test cases for pad.
PiperOrigin-RevId:
183846616
A. Unique TensorFlower [Tue, 30 Jan 2018 17:42:48 +0000 (09:42 -0800)]
Reenable 'constant' test.
PiperOrigin-RevId:
183845007
Shivani Agrawal [Tue, 30 Jan 2018 17:21:57 +0000 (09:21 -0800)]
Fixes broken link in documentation.
PiperOrigin-RevId:
183842485
Chris Leary [Tue, 30 Jan 2018 16:43:37 +0000 (08:43 -0800)]
[XLA] Plumb build options via local API.
* Break build options into their own translation unit for use from local client
and to mirror ExecutableRunOptions.
* Add some ToString()s to aid debugging.
* Add HLO graph generation regex to build options.
* Add SWIG type map for ExecutableBuildOptions.
Also fix a build issue occurring on some platforms with triangular_solve.
PiperOrigin-RevId:
183837856