ManHyuk [Sun, 27 May 2018 07:03:14 +0000 (16:03 +0900)]
Merge branch 'master' into fix_typo
Tatiana Shpeisman [Sat, 26 May 2018 00:06:29 +0000 (17:06 -0700)]
Merge pull request #19483 from markdryan/markdryan/fix-bfcallocator-align
Fix BFCAllocator::Extend alignment issues
Robin Richtsfeld [Fri, 25 May 2018 23:38:33 +0000 (01:38 +0200)]
Fix of issue #13164 (Merges #13382) (#16368)
* tf.gather int64 GPU, tf.gather_nd int32/int64 GPU, tf.scatter_nd int32 GPU
* Fix tf.gather test
Tatiana Shpeisman [Fri, 25 May 2018 22:52:39 +0000 (15:52 -0700)]
Merge pull request #19364 from Intel-tensorflow/agramesh/mkl_v14_fix2
[INTEL MKL] Upgrading to MKL DNN v.14
Tatiana Shpeisman [Fri, 25 May 2018 22:51:00 +0000 (15:51 -0700)]
Merge pull request #19540 from Intel-tensorflow/mkl_registration_fix
INTEL-MKL: Fix an issue related to MKL op registration
Michael Case [Fri, 25 May 2018 22:33:36 +0000 (15:33 -0700)]
Merge pull request #19558 from bzier/patch-1
Updated embedding column example
Michael Case [Fri, 25 May 2018 22:28:58 +0000 (15:28 -0700)]
Merge pull request #19544 from yongtang/
05182018-duplicate-import
Remove duplicate `from six import text_type` in upload_test_benchmarks.py
ankurtaly [Fri, 25 May 2018 21:43:38 +0000 (14:43 -0700)]
Merge pull request #19470 from ankurtaly/branch_197583446
Branch
197583446
Tatiana Shpeisman [Fri, 25 May 2018 19:49:54 +0000 (12:49 -0700)]
Merge pull request #19479 from Intel-tensorflow/concat_build_fix
INTEL-MKL: Fix build issue with old GCC version - MklConcat related
Brian Zier [Fri, 25 May 2018 17:49:57 +0000 (10:49 -0700)]
Updated embedding column example
The embedding column example had a comment that seemed to incorrectly relate to the indicator column. Updated the comment to reflect the embedding column info. Also updated the name of the dimension variable to be consistent with the example from a few lines earlier.
Asim Shankar [Fri, 25 May 2018 17:38:23 +0000 (10:38 -0700)]
Merge pull request #17331 from kbsriram/easy-nn-grads
C++ gradients: Fractional*Pool, Soft{Plus,Sign}
Michael Case [Fri, 25 May 2018 16:54:36 +0000 (09:54 -0700)]
Merge pull request #19532 from yongtang/
05242018-gemmlowp
Enable mirror.bazel.build link for gemmlowp
Michael Case [Fri, 25 May 2018 16:36:45 +0000 (09:36 -0700)]
Merge pull request #19553 from silvasean/patch-1
Fix broken link
Sean Silva [Fri, 25 May 2018 07:27:46 +0000 (00:27 -0700)]
Fix broken link
Also, as a side note, I found this link confusing. I was expecting it to take me to a list of pre-made estimators. Not a definition of what a pre-made estimator is. (maybe the glossary definition should link to a list of pre-made estimators?)
ted chang [Fri, 25 May 2018 06:21:46 +0000 (23:21 -0700)]
Fix cmake for MacOS (#17005)
This change address cmake build issues for MacOS.
Also fixes #14712
Yong Tang [Thu, 24 May 2018 05:02:16 +0000 (05:02 +0000)]
Remove whitelisted gemmlowp and mirror.bazel.build link has been enabled.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Michael Case [Thu, 24 May 2018 22:08:23 +0000 (15:08 -0700)]
Merge pull request #19538 from yongtang/
05242018-BUILD.bazel
Use BUILD.bazel for third_party packages
Guozhong Zhuang [Thu, 24 May 2018 20:30:18 +0000 (13:30 -0700)]
fix an registration issue for MKL_ML op
Michael Case [Thu, 24 May 2018 20:06:18 +0000 (13:06 -0700)]
Merge pull request #19508 from briannemsick/issue-18908
tf.keras: Fix plot_model for Sequential model
Michael Case [Thu, 24 May 2018 18:50:34 +0000 (11:50 -0700)]
Merge pull request #19429 from evision-ai/fix-nsync-on-android
fix a bug in built nsync android library
Michael Case [Thu, 24 May 2018 18:49:51 +0000 (11:49 -0700)]
Merge pull request #19472 from fbcotter/master
Corrected compute_gradient docstring
William D. Irons [Thu, 24 May 2018 18:49:05 +0000 (13:49 -0500)]
Fix ppc64le compile failure libpng (#19291)
Building Tensorflow on ppc64le fails with the message:
ERROR: /root/tensorflow/tensorflow/cc/BUILD:422:1: Linking of rule
'//tensorflow/cc:ops/control_flow_ops_gen_cc' failed (Exit 1)
bazel-out/host/bin/_solib_local/_U_S_Stensorflow_Scc_Cops_Scontrol_Uflow
_Uops_Ugen_Ucc___Utensorflow/libtensorflow_framework.so:
undefined reference to `png_init_filter_functions_vsx'
collect2: error: ld returned 1 exit status
This is fixed by adding powerpc specific files from the png archive
for @org_tensorflow//tensorflow:linux_ppc64le builds only.
Yong Tang [Thu, 24 May 2018 17:13:03 +0000 (17:13 +0000)]
Use BUILD.bazel for third_party packages
Both `BUILD` and `BUILD.bazel` could be used as the bazel project file
and `BUILD.bazel` is preferred (see https://github.com/bazelbuild/bazel/issues/4517#issuecomment-
360213750)
This fix changes generated BUILD in third_party packages to `BUILD.bazel`.
This will help avoid conflict with `BUILD` or `build` file/directory names
in third party packages.
For example, while working on 19461 I noticed that apache thrift package
consists of a `build` directory and that causes issues in case-insensitive
systems like macOS. This PR should help avoid such conflict issues.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Michael Case [Thu, 24 May 2018 16:52:27 +0000 (09:52 -0700)]
Merge pull request #19509 from yongtang/19498-doc-reduce_sum
Update docstring for tf.reduce_sum/reduce.
Yong Tang [Thu, 24 May 2018 05:00:30 +0000 (05:00 +0000)]
Enable mirror.bazel.build link for gemmlowp
The mirror.bazel.build link for gemmlowp was disabled
as it was not propagated before. This fix enables the
mirror.bazel.build link for gemmlowp and removes the
related TODO from it.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Michael Case [Thu, 24 May 2018 01:19:24 +0000 (18:19 -0700)]
Merge pull request #19512 from yongtang/
05232018-pylint
Fix pylint issue in the master branch
Yong Tang [Wed, 23 May 2018 23:53:53 +0000 (23:53 +0000)]
Additional pylint fix.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Wed, 23 May 2018 23:40:59 +0000 (23:40 +0000)]
Fix pylint issue in the master branch
The latest master branch has the following pylint failure that caused
`Ubuntu Sanity — Internal CI build failed`:
```
53 FAIL: Found 2 non-whitelited pylint errors:
54 tensorflow/contrib/cmake/tools/create_def_file.py:47: [C0301(line-too-long), ] Line too long (106/80)
55
56 tensorflow/contrib/cmake/tools/create_def_file.py:61: [C0301(line-too-long), ] Line too long (90/80)
```
This PR addresses the above issues.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Ankur Taly [Wed, 23 May 2018 22:50:17 +0000 (15:50 -0700)]
updated based on CL
197644290
Gunhan Gulsoy [Wed, 23 May 2018 22:49:35 +0000 (15:49 -0700)]
Expose stream executor namespace in cmake shared object. (#19415)
Instead of perftools::gputools expose stream executor namespace in cmake shared object.
Ankur Taly [Wed, 23 May 2018 21:34:23 +0000 (14:34 -0700)]
fixed DirectSessionWithTrackingAllocTest keyboard_arrow_right CostModelTest
Yong Tang [Wed, 23 May 2018 21:00:21 +0000 (21:00 +0000)]
Update tf.reduce_max and tf.reduce_min
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Wed, 23 May 2018 20:59:22 +0000 (20:59 +0000)]
Update docs for tf.reduce_prod and tf.reduce_mean
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Wed, 23 May 2018 20:54:45 +0000 (20:54 +0000)]
Update docstring for tf.reduce_sum
This fix tries to address the issue raised in 19498 where
the description:
```
If axis has no entries, all dimensions are reduced, and a tensor with a single element is returned.
```
does not match the current behavior. The current behavior (matches `np.sum` as well) is actually:
```
If axis is None, all dimensions are reduced, and a tensor with a single element is returned.
```
This fix fixes the discrepancy for tf.reduce_sum.
This fix fixes 19498.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Brian Nemsick [Wed, 23 May 2018 17:50:23 +0000 (10:50 -0700)]
Remove model.model in vis_utils plotting.
Michael Case [Wed, 23 May 2018 16:46:14 +0000 (09:46 -0700)]
Merge pull request #19502 from sdalbsoo/typo
Fix ovic typo
Sdalbsoo [Wed, 23 May 2018 15:16:54 +0000 (00:16 +0900)]
typo
Martin Wicke [Tue, 22 May 2018 23:27:20 +0000 (16:27 -0700)]
Merge pull request #17070 from yupbank/refactor-tree-variables
add model restore support for tree and forest variables
Yan Facai (颜发才) [Tue, 22 May 2018 23:16:01 +0000 (07:16 +0800)]
BUG: keras.callbacks.TensorBoard raises an exception for non_trainale_weights (#19148)
* TST: write_grads for non_trainable_weights
* BUG: bypass non_trainable_weights for write_grad
* CLN: factor out write_grad loop
Mark Ryan [Tue, 22 May 2018 18:23:14 +0000 (18:23 +0000)]
Fix BFCAllocator::Extend alignment issues
The BFCAllocator::Extend method used an incorrect harcoded alignment
of 32 in two different places. This led to alignment check assertions when
Extend was called, as all memory is now expected to be 64 byte aligned.
Signed-off-by: Mark Ryan <mark.d.ryan@intel.com>
Guozhong Zhuang [Tue, 22 May 2018 20:04:50 +0000 (13:04 -0700)]
fix a concat related build issue with old GCC version
Jayaram Bobba [Tue, 22 May 2018 19:03:24 +0000 (12:03 -0700)]
Setting default openmp settings for MKL kernels (#19136)
* Change inter op defaults when built with MKL to avoid thread oversubscription
* Bump up default mkl inter_op to be less conservative
* Added default OMP settings that are expected to give reasonable performance when using MKL kernels
* Query CPUID for determining number of hyperthreads per physical core on Intel 64 architectures
* Style fixes
* Buildifier and clang-format style fixes
* Removed use of setenv due to concerns about thread safety
Fergal Cotter [Tue, 22 May 2018 18:44:57 +0000 (19:44 +0100)]
Corrected compute_gradient docstring
After testing the tf.test.compute_gradient function with complex inputs,
noticed that the docstring specifies that the real/imag partial derivatives
are in the four corners of a larger matrix, when the result given indicates
they are instead interleaved.
Michael Case [Tue, 22 May 2018 18:06:19 +0000 (11:06 -0700)]
Merge pull request #19386 from smistad/fix-cmake
CMake: fix include error when building with PYTHON_BINDINGS off
Michael Case [Tue, 22 May 2018 18:02:22 +0000 (11:02 -0700)]
Merge pull request #19446 from sarahedkins/fix/python-feeding-function-error
fix(python): ValueError len(dataframe.columns) + 1
Ankur Taly [Tue, 22 May 2018 17:44:24 +0000 (10:44 -0700)]
Merge commit for internal changes
Akshay Agrawal [Tue, 22 May 2018 17:26:00 +0000 (10:26 -0700)]
Make init_scope preserve the inner device stack when lifting into a graph.
Eager execution doesn't implement device stacks and in particular it doesn't support device functions (which determine the device on a per-op basis), so in general it's not possible to do the same when lifting into the eager context.
PiperOrigin-RevId:
197583446
Michael Case [Tue, 22 May 2018 16:59:54 +0000 (09:59 -0700)]
Merge pull request #17025 from kr-ish/numpy_input_fn_none_check
Adds a check for shuffle=None in numpy_input_fn
Michael Case [Tue, 22 May 2018 16:56:49 +0000 (09:56 -0700)]
Merge pull request #19445 from yongtang/
05122018-link
Fix incorrect link for nvidia drivers
Michael Case [Tue, 22 May 2018 16:51:02 +0000 (09:51 -0700)]
Merge pull request #19462 from yongtang/
05212018-docs-math_ops
Fix discrepancies between doc and implementation for math_ops
Dan Moldovan [Tue, 22 May 2018 16:43:06 +0000 (09:43 -0700)]
Special case the 'dict' call, which trips other mechanisms for built-ins.
PiperOrigin-RevId:
197576297
Benjamin Kramer [Tue, 22 May 2018 16:08:06 +0000 (09:08 -0700)]
[TF:XLA] Fix xla_interpreter_device build
PiperOrigin-RevId:
197571618
A. Unique TensorFlower [Tue, 22 May 2018 15:18:11 +0000 (08:18 -0700)]
Contributing guidelines, style guide and README updates
PiperOrigin-RevId:
197564905
A. Unique TensorFlower [Tue, 22 May 2018 15:14:49 +0000 (08:14 -0700)]
Update calls to addPassesToEmitFile
PiperOrigin-RevId:
197564506
A. Unique TensorFlower [Tue, 22 May 2018 15:12:41 +0000 (08:12 -0700)]
Fix a couple of broken links in the Swift For TensorFlow page.
PiperOrigin-RevId:
197564254
A. Unique TensorFlower [Tue, 22 May 2018 15:02:39 +0000 (08:02 -0700)]
Automated g4 rollback of changelist
197527651
PiperOrigin-RevId:
197562826
Benjamin Kramer [Tue, 22 May 2018 14:06:08 +0000 (07:06 -0700)]
[XLA:TF] Run buildifier on llvm.BUILD
Buildifier recently started sorting load args
https://github.com/bazelbuild/buildtools/commit/
3ac5f85b22bc44820c041d0cacd3bc2ed54e7742 which causes diffs in the output.
PiperOrigin-RevId:
197556554
ManHyuk [Tue, 22 May 2018 13:43:40 +0000 (22:43 +0900)]
Fix typo (#19450)
A. Unique TensorFlower [Tue, 22 May 2018 12:50:34 +0000 (05:50 -0700)]
[XLA] Optimize ShapeTree<T>
This optimizes ShapeTree quite significantly. In particular this optimizes for the common case of querying/iterating, copying and moving ShapeTrees.
* Allocate all ShapeTreeNodes inside a single, owned, vector. This reduces the number of memory allocations and improves cache performance.
* Instead of storing children nodes as unique_ptrs, store them as indices into the owning container's vector. This allows cheap copy-construction (a std::vector POD copy) and doesn't change the fast path (dereferencing a pointer is just as fast as dereferencing a base + offset).
* Instead of a unique_ptr<Shape>, use a shared_ptr<Shape>. This removes a load of copy-construction overhead at the cost of a shared_ptr over a unique_ptr (one extra allocation).
* Instead of computing ShapeIndexes on-demand in the iterators/ForEach*, precompute them during construction time. This adds a few more bytes per ShapeTree, but now we can...
* ... store a std::pair<ShapeIndex, T> as the ShapeTreeNode's data element. This allows us to provide a std::pair<K,V>&, STL-like interface from iterators without going through any of the previous unique_ptr hacks around storage lifetimes.
* Because we no longer need to iterate from the beginning to build up the ShapeIndex, we can now offer a ::find() function to return an iterator for a ShapeIndex in O(K) time. As the iteration order is guaranteed to be pre-order, this can be used (and will be, later) to speed up the fast-path of mutating a subtree of a ShapeTree from tf2xla::ExtractSubBuffers.
* Similarly because we now have a very standard, cheap STL interface with no performance cliffs, we can hopefully improve ShapedBuffer's copy and move constructors to be cheaper.
PiperOrigin-RevId:
197548717
A. Unique TensorFlower [Tue, 22 May 2018 09:27:45 +0000 (02:27 -0700)]
internal change
PiperOrigin-RevId:
197533162
A. Unique TensorFlower [Tue, 22 May 2018 09:21:30 +0000 (02:21 -0700)]
batch_util.h is generally useful so moved to util/ from kernels/ where it will be included in the pip package.
PiperOrigin-RevId:
197532524
A. Unique TensorFlower [Tue, 22 May 2018 08:35:36 +0000 (01:35 -0700)]
convert Pow op into something that is more recognizable, so we can have further
optimizations.
PiperOrigin-RevId:
197527651
A. Unique TensorFlower [Tue, 22 May 2018 08:01:01 +0000 (01:01 -0700)]
Automated g4 rollback of changelist
197487461
PiperOrigin-RevId:
197523867
A. Unique TensorFlower [Tue, 22 May 2018 07:44:47 +0000 (00:44 -0700)]
Unifiy the cuda toolchain definition of gcc/nvcc and cuda-clang.
gcc >= 7 will change how it treats -pie [1]; passing -pie after -shared
on the command line is not possible any more; given that the legacy way to
configure flags in the gcc/nvcc toolchain does not allow control over where
the flags go or how to provide -pie only for linking of binaries, we can
prevent this from breaking in the future by also using the new feature
mechanism for gcc/nvcc.
In addition to moving the gcc-specific workarounds in the toolchain to
cuda_configure.bzl, document them, so we don't need to rediscover them in the
future.
[1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77464
PiperOrigin-RevId:
197522719
A. Unique TensorFlower [Tue, 22 May 2018 06:37:12 +0000 (23:37 -0700)]
Enable tpu.rewrite to work on XLA CPU/GPU backends.
PiperOrigin-RevId:
197517946
ManHyuk [Tue, 22 May 2018 03:46:23 +0000 (12:46 +0900)]
fix typo
Justin Lebar [Tue, 22 May 2018 03:41:26 +0000 (20:41 -0700)]
[XLA:GPU] Implement trivial (one-replica) cross-replica-sum on XLA:GPU.
Also fix the CPU implementation to work in the case when there are
multiple operands to the cross-replica-sum op.
PiperOrigin-RevId:
197506311
ManHyuk [Tue, 22 May 2018 03:42:04 +0000 (12:42 +0900)]
Merge branch 'master' into fix_typo
A. Unique TensorFlower [Tue, 22 May 2018 03:27:53 +0000 (20:27 -0700)]
Update scan benchmarks to have a range of 16K-128K iterations. As of https://github.com/tensorflow/tensorflow/commit/
5802096c267c805f6a69798aac10aefef759bb9f, TensorFlow Eager no longer exhibits quadratic behavior. The benchmark is still ~5x slower in eager mode vs. graph mode, and maybe slightly worse than linear:
n Graph Time (s) Eager Time (s) Ratio
-----------------------------------------------
16K 0.35 1.8 5.1
32K 0.64 3.6 5.6
64K 1.1 7.2 6.5
128K 2.4 14.8 6.2
PiperOrigin-RevId:
197505257
Michael Kuperstein [Tue, 22 May 2018 03:06:39 +0000 (20:06 -0700)]
Internal Change
PiperOrigin-RevId:
197503560
gdh1995 [Mon, 21 May 2018 06:23:32 +0000 (14:23 +0800)]
fix a bug of the android Makefile for nsync library
NDK doesn't support thread_local variables which require destructors,
so on Android it should use __thread instead.
Observations:
* ProtoBuf and other libraries are not using thread_local on Android.
* In Tensorflow, there's a "thread_local" in code about CUDA,
which should be safe enough.
More discussions are on https://github.com/android-ndk/ndk/issues/360 .
Michael Case [Tue, 22 May 2018 02:45:21 +0000 (19:45 -0700)]
Internal Change
PiperOrigin-RevId:
197501805
Asim Shankar [Tue, 22 May 2018 02:30:52 +0000 (19:30 -0700)]
s/tfe.GradientTape/tf.GradientTape/
s/tfe.enable_eager_execution/tf.enable_eager_execution/
PiperOrigin-RevId:
197500751
Akshay Modi [Tue, 22 May 2018 01:53:54 +0000 (18:53 -0700)]
Improvements to util/nest.py and data/util/nest.py
Changes:
- Add a cache for type -> is_sequence to speed up Flatten/IsSequence
- Update data/util/nest.py flatten to use C Flatten
Before:
entry {
name: "EagerLinearRegressionBenchmark.eager_train_cpu"
iters: 2000
wall_time: 1.
91852378845
extras {
key: "examples_per_sec"
value {
double_value: 66717.9634521
}
}
}
After:
entry {
name: "EagerLinearRegressionBenchmark.eager_train_cpu"
iters: 2000
wall_time: 1.
74479198456
extras {
key: "examples_per_sec"
value {
double_value: 73361.1806638
}
}
}
PiperOrigin-RevId:
197497854
Frank Chen [Tue, 22 May 2018 01:14:30 +0000 (18:14 -0700)]
Adds support for specifying a discovery_service_url (via either a parameter or an environment variable) within TPUClusterResolver
PiperOrigin-RevId:
197494335
Austin Anderson [Tue, 22 May 2018 00:45:22 +0000 (17:45 -0700)]
Split generated_examples test into multiple test targets
PiperOrigin-RevId:
197490872
Smit Hinsu [Tue, 22 May 2018 00:42:15 +0000 (17:42 -0700)]
Introduce an option to allocate CUDA unified memory
PiperOrigin-RevId:
197490523
Justin Lebar [Tue, 22 May 2018 00:34:56 +0000 (17:34 -0700)]
[XLA] Two minor style-guide fixups.
- Fix TODO(b/123) style.
- Use a value type rather than an rvalue reference for a "sink" param.
PiperOrigin-RevId:
197489597
Sarah Edkins [Tue, 22 May 2018 00:24:40 +0000 (20:24 -0400)]
fix(python): ValueError len(dataframe.columns) + 1
A. Unique TensorFlower [Tue, 22 May 2018 00:18:06 +0000 (17:18 -0700)]
Make the quantize_and_dequantize op use the full quantized range when possible.
PiperOrigin-RevId:
197487461
Petros Mol [Tue, 22 May 2018 00:15:39 +0000 (17:15 -0700)]
Improves documentation of labels and logits arguments in hinge loss methods .
PiperOrigin-RevId:
197487120
A. Unique TensorFlower [Tue, 22 May 2018 00:03:40 +0000 (17:03 -0700)]
Supports initializing an Interpreter with a direct ByteBuffer of nativeOrder()
that contains bytes content of a valid tflite model.
PiperOrigin-RevId:
197485253
Yong Tang [Mon, 21 May 2018 23:57:17 +0000 (23:57 +0000)]
Update docstring for tf.round as it support all real types.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Mon, 21 May 2018 23:36:55 +0000 (23:36 +0000)]
Update docstring as pow supports `float16` now
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Mon, 21 May 2018 23:29:32 +0000 (23:29 +0000)]
Fix discrepancies between doc and implementation for math_ops
In the docstring of math_ops, there are some discrepancies
between the supported data types and the actual implementation:
`abs` supports `float16`.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Saurabh Saxena [Mon, 21 May 2018 23:43:53 +0000 (16:43 -0700)]
Ensure that saving/restoring iterator in CheckpointInputPipelineHook is performed *after* the _DatasetInitializerHook has been run.
In the TPUEstimator the _DatasetInitializerHook is present in the
EstimatorSpec.training_hooks. Since these are executed after the `hooks`
passed to Estimator.train the input pipeline checkpointing hook fails
since it finds an uninitialized iterator.
PiperOrigin-RevId:
197482609
Alexandre Passos [Mon, 21 May 2018 23:37:17 +0000 (16:37 -0700)]
Fixes issue with gradient tape when asking for the gradient of an intermediate tensor.
PiperOrigin-RevId:
197481473
A. Unique TensorFlower [Mon, 21 May 2018 23:30:42 +0000 (16:30 -0700)]
Improve error message in tensor.cc when IsAligned() test fails
by logging offending ptr value.
PiperOrigin-RevId:
197480534
Igor Saprykin [Mon, 21 May 2018 23:26:11 +0000 (16:26 -0700)]
Support a better interface for the single option case in combinations.py.
If there's only one combination for combination-based tests, it doesn't have to
be a list.
PiperOrigin-RevId:
197479773
A. Unique TensorFlower [Mon, 21 May 2018 23:14:10 +0000 (16:14 -0700)]
Add arithmetic optimizer stage that removes LogicalNot that takes a comparison as input, i.e.
!(a == b) => a != b
!(a != b) => a == b
!(a < b) => a >= b
!(a <= b) => a > b
!(a > b) => a <= b
!(a >= b) => a < b
PiperOrigin-RevId:
197477959
A. Unique TensorFlower [Mon, 21 May 2018 23:13:28 +0000 (16:13 -0700)]
Expose partition_strategy option in embedding_lookup_unique
PiperOrigin-RevId:
197477853
Benoit Steiner [Mon, 21 May 2018 23:02:33 +0000 (16:02 -0700)]
Optimize more reductions
PiperOrigin-RevId:
197476067
Michael Case [Mon, 21 May 2018 22:55:46 +0000 (15:55 -0700)]
Internal Change.
PiperOrigin-RevId:
197475076
Sanjoy Das [Mon, 21 May 2018 22:49:14 +0000 (15:49 -0700)]
Extract out a MatrixMatrixBlockPanelEmitter::Dimensions struct; NFC
This gives me a convenient place to note that the m/k/n here are not the m/k/n
for the entire GEMM. I didn't rename m/k/n to mc/kc/nr since the latter seems
somewhat redundant to me -- we could read 'c as 'column' and 'r' as 'row', but
that's the only possibility?
This refactoring will also be useful when implementing GEPP on top of GEBP.
PiperOrigin-RevId:
197474137
Michael Case [Mon, 21 May 2018 22:48:11 +0000 (15:48 -0700)]
Merge pull request #19254 from yongtang/
05122018-compute_accidental_hits
Add additional shape validation to `compute_accidental_hits`
Allen Lavoie [Mon, 21 May 2018 22:43:20 +0000 (15:43 -0700)]
Remove object-based checkpointing probes from Python 3 tf.train.Saver "name not found" stack traces
PiperOrigin-RevId:
197473101
Jiri Simsa [Mon, 21 May 2018 22:32:45 +0000 (15:32 -0700)]
Disable flaky batch_dataset_op_test
PiperOrigin-RevId:
197471172
A. Unique TensorFlower [Mon, 21 May 2018 21:47:37 +0000 (14:47 -0700)]
Allow using DNN to only train the embeddings and using the tree model for the final prediction.
PiperOrigin-RevId:
197462585
Michael Case [Mon, 21 May 2018 21:38:18 +0000 (14:38 -0700)]
Merge pull request #19257 from yongtang/
05122018-QuantizedReluX
Shape validation of `max_features` in `QuantizedReluX`
Dimitris Vardoulakis [Mon, 21 May 2018 21:25:04 +0000 (14:25 -0700)]
[TF:XLA] Delete cumulative_total_size to simplify the DFS scheduler.
It's unclear why we would assign cumulative_total_size as the total size of a single HLO, and deleting it doesn't make a difference in practice.
PiperOrigin-RevId:
197458260