andrehentz [Fri, 13 Apr 2018 13:37:03 +0000 (06:37 -0700)]
Merge pull request #17313 from jin/patch-1
Include cstring in logging.cc for use of strrchr()
jjsjann123 [Fri, 13 Apr 2018 11:21:15 +0000 (04:21 -0700)]
updated installation instructions for Tensowflow-TensorRT integration (#18135)
* updated installation instructions for Tensowflow-TensorRT integration
* Minor format changes to clean it up.
* Adding the python symlink command for devel packages too.
* Forcing the symlink creation.
* Updating the sed command for docker parameterized build.
Sami Kama [Fri, 13 Apr 2018 01:07:50 +0000 (18:07 -0700)]
Replace tuple<int,int,int> for version info with a class in DnnSupportr::GetVersion() (#18434)
* Replace tuple<int,int,int> for version info with a class
* Removed clang-format modifications on non-edited code
* Update dnn.h
Update the comment as per request of reviewer
Russell Klopfer [Thu, 12 Apr 2018 20:01:25 +0000 (16:01 -0400)]
crf_decode fails when sequence_length is 0 (#17755)
* updating documentation
* crf_decode fails when sequence_length is 0
* fixing line length
* more pylint fixes
Yong Tang [Thu, 12 Apr 2018 20:01:05 +0000 (13:01 -0700)]
Add tensor support for num_spectrogram_bins in linear_to_mel_weight_matrix (#17404)
* Add tensor support for num_spectrogram_bins in linear_to_mel_weight_matrix
This fix tries to address the issue raised in 16553 where it was not possible
to provide num_spectrogram_bins as a tensor or placeholder for
linear_to_mel_weight_matrix.
The reason comes from the implementation of `_validate_arguments`
which requires num_spectrogram_bins to be a python.
However, the validation here is not necessary as `num_spectrogram_bins`
will be passed to `math_ops.linspace`, which performs the validation
anyway. The validation in `math_ops.linspace` is done in shape function
and in kernel's `Compute()`. For that it makes sense to remove the
validation of `num_spectrogram_bins` in `_validate_arguments` so that
the issue raised in 16553 could be addressed.
This fix adds a test case to cover the changes. Also, the
error case of `num_spectrogram_bins < 0` has already been covered in the
existing test case:
https://github.com/tensorflow/tensorflow/blob/
013a6c7b3112573ba4d932c8a22bfaf45f648c77/tensorflow/contrib/signal/python/kernel_tests/mel_ops_test.py#L149-L165
This fix fixes 16553.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add test case for tensor support of num_spectrogram_bins in mel_ops.linear_to_mel_weight_matrix
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add comment for removing validation of num_spectrogram_bins
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update docstring
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update test case for num_spectrogram_bins
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Remove unused constant_op import to pass sanity check
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Seyed Majid Azimi [Thu, 12 Apr 2018 20:00:33 +0000 (22:00 +0200)]
Update nn.py (#17247)
adding missing quantized_relu which was missing before.
David Norman [Thu, 12 Apr 2018 20:00:12 +0000 (21:00 +0100)]
Allow for devices which have F16, no F64, no Complex (#17473)
Yihong Wang [Thu, 12 Apr 2018 19:59:48 +0000 (12:59 -0700)]
Link to gcc_s and gcc if compiler is GCC version 5 (#17849)
When using cmake and GCC 5.4 to build tensorflow in Ubuntu 16.04,
the following error message would show up when loading
_pywrap_tensorflow_internal.so:
```
_pywrap_tensorflow_internal.so: undefined symbol: __cpu_model
```
The root cause is the same to this issue:
https://github.com/tensorflow/tensorflow/issues/9593
Signed-off-by: Yihong Wang <yh.wang@ibm.com>
Quanlong [Thu, 12 Apr 2018 19:59:26 +0000 (03:59 +0800)]
Latest nngraph cannot build with Hexagon SDK 3.0 (#17963)
* fix: latest nngraph cannot build with Hexagon SDK 3.0
* Fix typo
Shaoning Zeng [Thu, 12 Apr 2018 19:58:59 +0000 (03:58 +0800)]
give some writing updates to tensorflow/contrib/slim/README.md (#18259)
* add missed right bracket in ### Scopes
* change one , to . in ### Scopes
* refine one sentence
Wai Hon Law [Thu, 12 Apr 2018 19:58:44 +0000 (12:58 -0700)]
Change --output_png to --output_image (#18273)
The argument is incorrect.
When running the given command, we get
```
E tensorflow/examples/wav_to_spectrogram/main.cc:54] Unknown argument
--output_png=/tmp/spectrogram.png
```
TESTED:Rerun the updated command and verify that the flag is correct.
```
bazel run tensorflow/examples/wav_to_spectrogram:wav_to_spectrogram --
--input_wav=/tmp/speech_dataset/happy/ab00c4b2_nohash_0.wav
--output_image=/tmp/spectrogram.png
```
brett koonce [Thu, 12 Apr 2018 19:58:36 +0000 (12:58 -0700)]
contrib/autograph: minor spelling tweaks (#18284)
Yan Facai (颜发才) [Thu, 12 Apr 2018 19:58:22 +0000 (03:58 +0800)]
remove the misleading n_class information (#18305)
* DOC: modify the misleading n_class info
* DOC: add suggested fix
Yong Tang [Thu, 12 Apr 2018 19:58:04 +0000 (12:58 -0700)]
Fix WARNING in BatchNormalization (#18315)
The keep_dims for reduce_mean has been deprecated and replaced
with keepdims. This casues the following WARNING in BatchNormalization:
```
normalization.py:584: calling reduce_mean (from tensorflow.python.ops.math_ops) with keep_dims is deprecated and will be removed in a future version.
Instructions for updating:
keep_dims is deprecated, use keepdims instead
```
This fix fixes the warning in BatchNormalization.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
brett koonce [Thu, 12 Apr 2018 19:57:48 +0000 (12:57 -0700)]
contrib: minor spelling tweaks (#18330)
* contrib: minor spelling tweaks
* Fix lint error
Rholais Lii [Thu, 12 Apr 2018 19:57:39 +0000 (03:57 +0800)]
Reorder section `Using SavedModel with Estimators` (#18412)
Outputs should be specified before performing an export.
Loo Rong Jie [Thu, 12 Apr 2018 19:57:26 +0000 (03:57 +0800)]
[MSVC] Remove -D__VERSION__ flag and implement tf_compiler_version properly (#18445)
Loo Rong Jie [Thu, 12 Apr 2018 19:36:52 +0000 (03:36 +0800)]
Replace all COMPILER_MSVC to _MSC_VER and _WIN32 accordingly (#18448)
* Replace all COMPILER_MSVC to _MSC_VER and _WIN32 accordingly
* One more ARRAYSIZE to TF_ARRAYSIZE
* Delete non-existing include
andrehentz [Thu, 12 Apr 2018 18:05:30 +0000 (11:05 -0700)]
Merge pull request #17123 from hovhannesgithub/div-sub-broadcasting
Add broadcasting functionality for Div and Sub ops.
Hovhannes Harutyunyan [Thu, 12 Apr 2018 06:54:41 +0000 (10:54 +0400)]
Fixe merge issue
Hovhannes Harutyunyan [Thu, 12 Apr 2018 06:10:09 +0000 (10:10 +0400)]
Merge branch 'master' into div-sub-broadcasting
Hovhannes Harutyunyan [Thu, 12 Apr 2018 06:04:40 +0000 (10:04 +0400)]
Remove redefined BroadcastDiv function
Yanbo Liang [Thu, 12 Apr 2018 01:42:50 +0000 (18:42 -0700)]
Correct argument doc for BasicLSTMCell.call (#16554)
* Correct argument doc for BasicLSTMCell.call
* change self._num_units to num_units.
fo40225 [Thu, 12 Apr 2018 01:41:48 +0000 (09:41 +0800)]
fix tf.GIT_VERSION always 'unknown' on windows cmake build (#16730)
Yong Tang [Thu, 12 Apr 2018 01:34:20 +0000 (18:34 -0700)]
Enhancement with deprecated_argument_lookup (#17527)
* Enhancement with deprecated_argument_lookup
The tf.losses.cosine_distance deprecated dim and switched
to axis. This fix adds the enhancement of using
deprecated_argument_lookup, which is used in all other
arguments deprecations.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add missing import
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Seungwoo Choi (Biggie) [Thu, 12 Apr 2018 01:33:41 +0000 (10:33 +0900)]
Replace wrong variable (#17738)
Michal Turek [Thu, 12 Apr 2018 01:32:10 +0000 (03:32 +0200)]
Add comment to examples to prevent resource leaks (#17820)
Issue #17374
tamimaddari82 [Thu, 12 Apr 2018 01:28:18 +0000 (09:28 +0800)]
Add parallel implementation of CTC greedy decoder (#17982)
brett koonce [Thu, 12 Apr 2018 01:27:16 +0000 (18:27 -0700)]
contrib/image: minor spelling tweaks (#18162)
Yong Tang [Thu, 12 Apr 2018 01:26:12 +0000 (18:26 -0700)]
Check input dimension for contrib.layers.conv2d/conv3d (#18251)
* Check input dimension for contrib.layers.conv2d/conv3d
This fix tries to fix the issue raised in 14583 where
the input dimension was not checked for contrib.layers.conv2d/conv3d
and contrib.slim.conv2d/conv3d.
The issue was that conv2d/conv3d were just aliases of
convolution. This fix wrap the conv2d/conv3d with the input
dimension check so that incorrect usage will return ValueError.
This fix fixes 14583.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add test case for conv2d/conv3d shape check
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Fix impacted tests.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update convolution instead of adding _convolution,
based on review feedback
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add convolution1d and additional update
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
jinghuangintel [Thu, 12 Apr 2018 01:20:55 +0000 (18:20 -0700)]
added missing shapefn to several operators (#18298)
Ivan Zhang [Thu, 12 Apr 2018 01:18:01 +0000 (21:18 -0400)]
Fix typo in error message (#18319)
David Norman [Thu, 12 Apr 2018 01:09:31 +0000 (02:09 +0100)]
Disable int64 test for backends which don't support it (#18344)
ImSheridan [Thu, 12 Apr 2018 01:09:17 +0000 (09:09 +0800)]
Fix broken links in /extend/language_bindings (#18346)
ngc92 [Thu, 12 Apr 2018 01:08:42 +0000 (03:08 +0200)]
unified flip_* and random_flip_* functions (#18364)
Vadim Markovtsev [Thu, 12 Apr 2018 01:08:00 +0000 (03:08 +0200)]
Replace print with logging (#18392)
Mahmoud Abuzaina [Thu, 12 Apr 2018 01:04:26 +0000 (18:04 -0700)]
Fixing non-mkl builds (#18401)
Yong Tang [Thu, 12 Apr 2018 01:01:38 +0000 (18:01 -0700)]
Add negative axis support for tf.manip.roll (#18409)
* Add negative axis support for tf.manip.roll
This fix tries to support negative axis for tf.manip.roll.
The tf.manip.roll is supposed to be compatible with numpy.roll
which does support negative axis.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add test case for negative axis support for tf.manip.roll
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add axis check so that negative axis is within the range
Negative axis should be 0 <= axis + dims < dims
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add additional test cases
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Fix pylint issue
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Thu, 12 Apr 2018 00:51:41 +0000 (17:51 -0700)]
Add deprecated_args decoration to expand_dims (#18419)
* Add deprecated_args decoration to expand_dims
This fix adds deprecated_args decoration to expand_dims
as `dims` has been deprecated and in favor of `axis`
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Enhance deprecated args with deprecation.deprecated_argument_lookup
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Thu, 12 Apr 2018 00:50:45 +0000 (17:50 -0700)]
Imporve shape function of RandomUniformInt (#18420)
* Imporve shape function of RandomUniformInt
The input of `minval` and `maxval` of `RandomUniformInt`
should be scalar though it is not checked in the shape
function. This fix improves the shape function with the
rank check, and adds test case for it.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add test case for maxval and minval for RandomUniformInt
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Thu, 12 Apr 2018 00:49:47 +0000 (17:49 -0700)]
Update docs of reduce_max/reduce_min for real numeric type (#18422)
Both reduce_max and reduce_min only work for real numeric type
as complex numbers do not apply. This fix update the docs
with `numeric type` -> `real numeric type`.
Note that the current kernel registration in
reduction_ops_max.cc and reduction_ops_min.cc
use `TF_CALL_REAL_NUMBER_TYPES` so it is good.
The op registraton for Max and Min inside math_ops.cc should be `.Attr("T: realnumbertype")` instead of `numbertype`.
However, such a change will break API compatibility so leave it alone.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
bhavani-subramanian [Wed, 11 Apr 2018 22:10:38 +0000 (15:10 -0700)]
[INTEL MKL] Skip special nodes inserted by TF and MKL (#18077)
* Skip special nodes inserted by TF. This fixes TFDO-178 JIIRA issue.
* Added a comment about skipping nodes with an /_ in them.
* Stripped trailing whitespace.
* Wrapped code such that it is executed only when INTEL_MKL is defined.
Billy Lamberta [Wed, 11 Apr 2018 21:34:32 +0000 (14:34 -0700)]
Merge pull request #18340 from JChord/master
Docs: Fix 'Unable to find source java class'
Yong Tang [Wed, 11 Apr 2018 20:35:51 +0000 (13:35 -0700)]
Update tf.contrib.metrics with deprecations (#18335)
* Update tf.contrib.metrics with deprecations
This fix updates tf.contrib.metrics.streaming_mean_absolution(relative/squared)_error
with deprecation notices as they have been replaces with tf.metrics.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update streaming_mean_relative_error
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update deprecation notice for streaming_root_mean_squared_error
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Fix the deprecation message.
* Fix pylint `Line too long (81/80)` issue
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Rajendra arora [Wed, 11 Apr 2018 19:49:31 +0000 (01:19 +0530)]
Updating tensorboard link in Readme.md (#18161)
Billy Lamberta [Wed, 11 Apr 2018 19:07:59 +0000 (12:07 -0700)]
Merge pull request #18430 from lamberta/fix-docs-using-tpu
Docs: Clarify using_tpu.md
Billy Lamberta [Wed, 11 Apr 2018 19:05:39 +0000 (12:05 -0700)]
Tidy up doc for rebuild project
Qianli Scott Zhu [Wed, 11 Apr 2018 18:27:12 +0000 (11:27 -0700)]
Merge pull request #18426 from qlzh727/branch_192461382
Branch
192461382
Billy Lamberta [Wed, 11 Apr 2018 17:57:30 +0000 (10:57 -0700)]
Docs: Clarify using_tpu.md
Scott Zhu [Wed, 11 Apr 2018 16:54:10 +0000 (09:54 -0700)]
Merge commit for internal changes
Nupur Garg [Wed, 11 Apr 2018 16:34:51 +0000 (09:34 -0700)]
Fixing dependencies.
PiperOrigin-RevId:
192461382
Michael Case [Wed, 11 Apr 2018 16:34:44 +0000 (09:34 -0700)]
Fix Windows GPU TensorFlow Bazel builds.
The configure.py script will error out on Windows GPU builds due
to NCCL attempted to be configured (and is currently Linux only).
PiperOrigin-RevId:
192461362
A. Unique TensorFlower [Wed, 11 Apr 2018 15:30:18 +0000 (08:30 -0700)]
Temporarily remove prelu from generated_examples_zip_test
PiperOrigin-RevId:
192453411
A. Unique TensorFlower [Wed, 11 Apr 2018 15:11:50 +0000 (08:11 -0700)]
Remove unused former source of tensorflow.org/tutorials/image_retraining.
The source of https://tensorflow.org/tutorials/image_retraining has moved from
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/docs_src/tutorials
to https://github.com/tensorflow/hub/tree/master/docs/tutorials
because of its use of TensorFlow Hub.
This change replaces the now-defunct version with a pointer to the new
location, in order to avoid dead code.
PiperOrigin-RevId:
192451570
A. Unique TensorFlower [Wed, 11 Apr 2018 14:47:26 +0000 (07:47 -0700)]
Fixing propagation of minmax info on constant gather ops.
PiperOrigin-RevId:
192448922
Benjamin Kramer [Wed, 11 Apr 2018 09:16:25 +0000 (02:16 -0700)]
[TF:XLA] Mark oom_test as optonly, it's really slow when compiled without optimization.
PiperOrigin-RevId:
192420481
Nick Felt [Wed, 11 Apr 2018 06:44:12 +0000 (23:44 -0700)]
Create FileWriter <-> tf.contrib.summary compatibility layer
This provides an implementation of FileWriter, activated by passing in a `session` parameter to the constructor, that is backed by session.run'ing graph ops that manipulate a tf.contrib.summary.create_file_writer() instance. Because tf.contrib.summary.SummaryWriters are backed by shared resources in the graph, this makes it possible to have a FileWriter and a tf.contrib.summary.SummaryWriter that both write to the same events file.
This change includes some related smaller changes:
- Factors out training_utils.py into a separate target to avoid a cyclic dep
- Moves contrib/summary/summary_ops.py to python/ops/summary_ops_v2.py
- Adds SummaryWriter.init(), .flush(), and .close() op-returning methods
- Changes create_file_writer() `name` arg to default to logdir prefixed by `logdir:` so shared resources are scoped by logdir by default
- Fixes a bug with tf.contrib.summary.flush() `writer` arg
- Makes create_file_writer()'s max_queue arg behave as documented
- Adds more testing for existing tf.contrib.summary API
PiperOrigin-RevId:
192408079
A. Unique TensorFlower [Wed, 11 Apr 2018 05:44:36 +0000 (22:44 -0700)]
[XLA] Fix the size of data buffer for sparse literals.
PiperOrigin-RevId:
192404543
A. Unique TensorFlower [Wed, 11 Apr 2018 05:29:13 +0000 (22:29 -0700)]
[XLA] Redesign: implement ComputeHost.
Also support convert from/to proto for ComputeHost.
PiperOrigin-RevId:
192403660
A. Unique TensorFlower [Wed, 11 Apr 2018 04:42:14 +0000 (21:42 -0700)]
[XLA] Redesgin: implement and test Gather, Conditional.
Also support convert from/to proto for Gather.
PiperOrigin-RevId:
192400659
A. Unique TensorFlower [Wed, 11 Apr 2018 03:48:57 +0000 (20:48 -0700)]
[XLA] Redesign: implment and test CrossReplicaSum.
PiperOrigin-RevId:
192397189
A. Unique TensorFlower [Wed, 11 Apr 2018 02:44:00 +0000 (19:44 -0700)]
Update ops-related pbtxt files.
PiperOrigin-RevId:
192392702
Justin Lebar [Wed, 11 Apr 2018 02:34:54 +0000 (19:34 -0700)]
[StreamExecutor] Merge StreamExecutor's and XLA's StatusOr classes.
StatusOr is a...complicated class to write. It's really not good to
have two copies of it. They've diverged (the XLA one is more
sophisticated), and this may be causing upstream build problems with
gcc6.
PiperOrigin-RevId:
192392111
A. Unique TensorFlower [Wed, 11 Apr 2018 02:31:05 +0000 (19:31 -0700)]
[XLA] Redesign: implement and test BatchNormXXX.
PiperOrigin-RevId:
192391748
Asim Shankar [Wed, 11 Apr 2018 02:20:58 +0000 (19:20 -0700)]
experimental C API: Fix compilation failure in Windows.
The functions added in
https://github.com/tensorflow/tensorflow/commit/
be917027e37c5e8f21f6ba07f24bdbf072cf6dfd
are temporary, and their existence breaks compilation in MSVC because of
https://docs.microsoft.com/en-us/cpp/c-language/maximum-string-length
and
https://docs.microsoft.com/en-us/cpp/error-messages/compiler-errors-1/compiler-error-c2026
So just disabling it in Windows for now.
PiperOrigin-RevId:
192391164
Allen Lavoie [Wed, 11 Apr 2018 02:01:33 +0000 (19:01 -0700)]
Checkpointable: remove colocation constraints from restore ops
Mystery solved thanks to log_device_placement.
PiperOrigin-RevId:
192389574
Francois Chollet [Wed, 11 Apr 2018 01:50:58 +0000 (18:50 -0700)]
Remove BN workaround for resource variable gradients bug that was recently fixed.
PiperOrigin-RevId:
192388867
Loo Rong Jie [Wed, 11 Apr 2018 01:51:10 +0000 (09:51 +0800)]
Add missing TF_ATTRIBUTE_WEAK for MSVC (#18303)
Michael Case [Wed, 11 Apr 2018 01:44:13 +0000 (18:44 -0700)]
Merge changes from github.
PiperOrigin-RevId:
192388250
Francois Chollet [Wed, 11 Apr 2018 01:41:37 +0000 (18:41 -0700)]
Fix for users who were passing `Dimension` type as `units` arg in `Dense`.
PiperOrigin-RevId:
192387984
Yong Tang [Wed, 11 Apr 2018 01:38:06 +0000 (18:38 -0700)]
Fix incorrect math equation renderings in random_fourier_features.py (#18367)
* Fix incorrect math equation renderings in random_fourier_features.py
This fix fixes incorrect math equation renderings for markdown
in random_fourier_features.py. The issue is that
"```" backtick should not be added when mathjax quote is used ("\\(" or "$$").
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Additional fix.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* MathJax fixes
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Fix pylint errors
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Wed, 11 Apr 2018 01:37:35 +0000 (18:37 -0700)]
Fix code block rendering issue in adding_an_op.md (#18368)
* Fix code block rendering issue in adding_an_op.md
In adding_an_op.md, html code was used in markdown for code blocks.
However, this does not work very well as some of the code blocks
includes incorrect rendering.
This fix converts html into "```c++" (backticks) so that the rendering
could be fixed.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Fix additional html code
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Fix lang-cpp issue
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Further clean up
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Maciej [Wed, 11 Apr 2018 01:37:03 +0000 (20:37 -0500)]
typo and readability fixes in CPU section (#18370)
Fixed a typo in the Tuning MKL section, and modified punctuation for intra_op_parallelism_threads section for easier readability.
Allen Lavoie [Wed, 11 Apr 2018 01:04:20 +0000 (18:04 -0700)]
TFTS: De-flake the LSTM test
Disabling the value-based check for now. Hopefully the shapes are deterministic.
PiperOrigin-RevId:
192383553
Allen Lavoie [Wed, 11 Apr 2018 00:57:19 +0000 (17:57 -0700)]
Forward the status from LookupResource to GetInputTensorFromVariable rather than returning a generic error status
PiperOrigin-RevId:
192382499
A. Unique TensorFlower [Wed, 11 Apr 2018 00:49:51 +0000 (17:49 -0700)]
[XLA] Redesign: implement Conj.
PiperOrigin-RevId:
192381481
A. Unique TensorFlower [Wed, 11 Apr 2018 00:47:13 +0000 (17:47 -0700)]
[XLA] Redesign: implement SliceInDim.
PiperOrigin-RevId:
192381080
A. Unique TensorFlower [Wed, 11 Apr 2018 00:44:48 +0000 (17:44 -0700)]
[XLA] Redesign: implement ReduceAll.
PiperOrigin-RevId:
192380688
A. Unique TensorFlower [Wed, 11 Apr 2018 00:41:56 +0000 (17:41 -0700)]
[XLA] Redesign: implement and test SelectAndScatter.
PiperOrigin-RevId:
192380121
Austin Anderson [Wed, 11 Apr 2018 00:37:53 +0000 (17:37 -0700)]
Update LogToSTDErr for TF Lite usage
PiperOrigin-RevId:
192379483
Jacques Pienaar [Tue, 10 Apr 2018 23:47:08 +0000 (16:47 -0700)]
Add types to error message in case of mismatch. NFC.
PiperOrigin-RevId:
192370979
Michael Kuperstein [Tue, 10 Apr 2018 23:45:19 +0000 (16:45 -0700)]
Add a thread-safe producer-consumer queue.
PiperOrigin-RevId:
192370670
Alexandre Passos [Tue, 10 Apr 2018 23:35:58 +0000 (16:35 -0700)]
Fix issue with gradients of resource variables in cond.
PiperOrigin-RevId:
192369091
A. Unique TensorFlower [Tue, 10 Apr 2018 23:32:05 +0000 (16:32 -0700)]
[XLA] Redesign: implement and test ReduceWindow.
PiperOrigin-RevId:
192368401
Amit Patankar [Tue, 10 Apr 2018 23:36:13 +0000 (16:36 -0700)]
Merge pull request #18407 from av8ramit/fix_sed_cmd
Updating the sed command for docker parameterized build.
A. Unique TensorFlower [Tue, 10 Apr 2018 23:31:46 +0000 (16:31 -0700)]
Update programmers guide
PiperOrigin-RevId:
192368335
A. Unique TensorFlower [Tue, 10 Apr 2018 23:26:05 +0000 (16:26 -0700)]
Supporting FakeQuant num_bits and getting the fake quant op matching tensorflow.
PiperOrigin-RevId:
192367307
Chris Leary [Tue, 10 Apr 2018 23:12:19 +0000 (16:12 -0700)]
[XLA] GRPC service definition.
PiperOrigin-RevId:
192364932
Francois Chollet [Tue, 10 Apr 2018 23:11:38 +0000 (16:11 -0700)]
Fix `nn` module RNN namespace issues.
PiperOrigin-RevId:
192364808
Amit Patankar [Tue, 10 Apr 2018 23:10:13 +0000 (16:10 -0700)]
Updating the sed command for docker parameterized build.
Mingxing Tan [Tue, 10 Apr 2018 22:59:39 +0000 (15:59 -0700)]
Allow passing allow_custom_ops for toco_convert.
PiperOrigin-RevId:
192362688
annarev [Tue, 10 Apr 2018 23:01:23 +0000 (16:01 -0700)]
Merge pull request #18399 from av8ramit/fix_docker_link
Forcing the symlink creation.
Alexandre Passos [Tue, 10 Apr 2018 22:49:03 +0000 (15:49 -0700)]
Cleaning up _distributed_apply now the device policy is unnecessary
PiperOrigin-RevId:
192360913
Akshay Agrawal [Tue, 10 Apr 2018 22:48:15 +0000 (15:48 -0700)]
In `get_variable`, nest the choice to use `ResourceVariable` under an
`init_scope`.
This makes sure that, when executing eagerly, calls to `get_variable` in a
`defun`-compiled function retrieve `ResourceVariable`s instead of `Variables`.
PiperOrigin-RevId:
192360775
A. Unique TensorFlower [Tue, 10 Apr 2018 22:45:37 +0000 (15:45 -0700)]
Small changes to testing code, plus a new binary to check diff from command line.
PiperOrigin-RevId:
192360373
A. Unique TensorFlower [Tue, 10 Apr 2018 22:39:37 +0000 (15:39 -0700)]
[XLA] Redesign: implement and test Conv.
PiperOrigin-RevId:
192359226
A. Unique TensorFlower [Tue, 10 Apr 2018 22:23:05 +0000 (15:23 -0700)]
Allow negative feature values in computation for `sum` combiner.
PiperOrigin-RevId:
192355950
A. Unique TensorFlower [Tue, 10 Apr 2018 22:01:49 +0000 (15:01 -0700)]
Add quantized LogSoftmax.
PiperOrigin-RevId:
192352432
Amit Patankar [Tue, 10 Apr 2018 21:59:23 +0000 (14:59 -0700)]
Forcing the symlink creation.
A. Unique TensorFlower [Tue, 10 Apr 2018 21:51:54 +0000 (14:51 -0700)]
Run EvaluateNodes for ModelPrunerTest_StopGradientPruning. Also updated the test fixture to inherit from GrapplerTest.
PiperOrigin-RevId:
192350828