Gunhan Gulsoy [Tue, 15 May 2018 17:41:38 +0000 (10:41 -0700)]
Add IBM ppc64le build to README. (#19285)
* Add IBM ppc64le build to README.
* ppc64le -> ppc64le CPU
Yong Tang [Tue, 15 May 2018 00:58:36 +0000 (17:58 -0700)]
Add tf.regex_match for regex match support (#19160)
* Add tf.regex_match for regex match support
This fix tries to address the issue raised in 18264.
Currently tf.regex_replace has already been supported
though there was no regex match support.
This fix adds the tf.regex_match support in a similiar
pattern as tf.regex_replace.
This fix fixes 18264.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update BUILD file for the tf.regex_match kernel
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Register RegexMatch ops
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add test cases for tf.regex_match
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update api_defs
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update API golden
update with:
```
bazel-bin/tensorflow/tools/api/tests/api_compatibility_test
--update_goldens True
```
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Expose regex_full_match in tf.strings namespace
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update golden API
```
bazel-bin/tensorflow/tools/api/tests/api_compatibility_test
--update_goldens True
```
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Tue, 15 May 2018 00:57:23 +0000 (17:57 -0700)]
Improve shape function of `tf.image.draw_bounding_boxes` (#19237)
* Improve shape function of `tf.image.draw_bounding_boxes`
The `tf.image.draw_bounding_boxes` requires `boxes` to be
3-D shape though there was no check on shape function.
This fix improves the shape function by restricting the
boxes to 3-D.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add additional check to make sure boxes shape
ends with 4 ([batch, num_bounding_boxes, 4])
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Address review feedback with addtional shape checks.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add unit tests
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Mon, 14 May 2018 18:35:40 +0000 (11:35 -0700)]
Add int64 support for output_shape of tf.nn.conv3d_transpose (#19248)
* Add int64 support for output_shape of tf.nn.conv3d_transpose
This fix tries to address the issue raised in 18887 where
the output_shape of tf.nn.conv3d_transpose only support
int32 data types. The support of int64 has been added in this PR
with test case covered.
This fix fixes 18887.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update op registration for Conv3DBackpropInputV2
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add test case for int64 support of output_shape with tf.nn.conv3d_transpose
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update test case with both int32 and int64
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Fix pylint issue
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yilei (Dolee) Yang [Mon, 14 May 2018 18:34:54 +0000 (11:34 -0700)]
Fix links on the community/swift page. (#19230)
They were broken rendered on https://www.tensorflow.org/community/swift.
Yong Tang [Mon, 14 May 2018 18:12:32 +0000 (11:12 -0700)]
Fix warning caused by squeeze_dims (#19227)
The `squeeze_dims` in `tf.squeeze` has been deprecated
in favor of `axis`. This fix fixes the `squeeze_dims`
in text_classification_cnn.py so that the warning
could be removed.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Mon, 14 May 2018 18:10:27 +0000 (11:10 -0700)]
Add complex support for tf.segment_mean (#19225)
* Add complex support for tf.segment_mean
While using tf.segment_mean I noticed that it does not
have the complex support like tf.segment_sum. I think it
makes sense to support complex for it. This fix adds the
complex support for tf.segment_mean.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Add test cases for complex support with tf.segment_mean
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yong Tang [Mon, 14 May 2018 18:09:30 +0000 (11:09 -0700)]
Fix misleading cupti.h error message (#19224)
This fix tries to address the issue raised in 19223
where the cupti.h eror message was misleading. The following error:
```
Cuda Configuration Error: Cannot find cupti.h under /usr/local/cuda-9.0
```
is not the true patch searched.
This fix updates the bzl file to print out the
complete searched paths when error occurs:
```
Cuda Configuration Error: Cannot find cupti.h under /usr/local/cuda-9.0/extras/CUPTI/include/, /usr/local/cuda-9.0/include/cuda/CUPTI/
```
This fix fixes 19223.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Jan Zikes [Mon, 14 May 2018 16:03:34 +0000 (18:03 +0200)]
Enable OrderedEnqueuer from keras in tf.keras. (#19183)
Paul Van Eck [Mon, 14 May 2018 15:27:42 +0000 (08:27 -0700)]
Make op unique name generation case insensitive (#18413)
* Make op unique name generation case insensitive
Unique name generation for operations depends on checking a dict for
names currently in use. This commit makes it so that the names stored in
this dict are always lowercase so that we can check if a name already
exists regardless of the capitalization.
This helps in filesystems where file paths are case insensitive and
tensor dumps (like with tfdbg) try to follow directory structures that
correspond to the tensor names. If two tensors have names with the same
spelling, but different capitalizations, then this can lead to unintended
side effects/errors on these case-insensitive file systems.
* Change variable name to match unique_name
* Adjust op names to fix tests
David Norman [Sun, 13 May 2018 02:11:23 +0000 (19:11 -0700)]
Allow for disabling of 2 tests (#18208)
AG Ramesh [Sun, 13 May 2018 01:35:11 +0000 (18:35 -0700)]
Fix for crash in mkl_layout_pass_test (#19107)
Gunhan Gulsoy [Fri, 11 May 2018 22:44:14 +0000 (15:44 -0700)]
Update how build statuses and artifacts are demoed in README.md (#19232)
Yong Tang [Fri, 11 May 2018 16:48:32 +0000 (09:48 -0700)]
Improve the shape function for ParameterizedTruncatedNormal (#19215)
The parameters of ParameterizedTruncatedNormal should
be 0-D or 1-D, which is checked in ther kernel functions.
There is no check in the shape function of the ops.
This fix improves the shape function and checks the
parameters of ParameterizedTruncatedNormal whever possible.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
David Norman [Fri, 11 May 2018 16:40:34 +0000 (09:40 -0700)]
Allow for disabling of some tests (#19202)
Achal Shah [Fri, 11 May 2018 00:43:30 +0000 (17:43 -0700)]
Fix cublas wrap macro for cublasGemmBatchedEx (#19210)
ManHyuk [Thu, 10 May 2018 21:11:43 +0000 (06:11 +0900)]
Fix typo (#19106)
* fix typo
Martin Wicke [Thu, 10 May 2018 19:20:12 +0000 (12:20 -0700)]
Merge pull request #18759 from filipefilardi/master
Add Pull Request Template
Yong Tang [Thu, 10 May 2018 18:38:04 +0000 (11:38 -0700)]
Use "```" (backtick) for code blocks in adding_an_op.md (#19187)
* Use "```" (backtick) for code blocks in adding_an_op.md
In adding_an_op.md, most of the code blocks uses "```" (backtick)
and annotations are added automatically. Though there was one
place where the code block are done with manual html code. This
is really error-prune and hard to change if there is an update
in the future.
This fix converts to "```c++" (backticks) so that it is easy
to maintain in the future.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Fix extra `\` at the beginning of the block
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Update adding_an_op.md
add new lines where the <br/> tags were.
Mahmoud Abuzaina [Thu, 10 May 2018 18:37:19 +0000 (11:37 -0700)]
Fixing util_cuda_kernel_helper_test_gpu when building with MKL enabled (#19185)
P-Hidringer [Thu, 10 May 2018 18:36:57 +0000 (03:36 +0900)]
Fix default value for parameter 'prefix' in slim.tfexample_decoder.BoundingBox.__init__ (#19190)
Ben Barsdell [Thu, 10 May 2018 18:06:01 +0000 (11:06 -0700)]
Add GPU support for float16 batched matmul (#18436)
* Add GPU support for float16 batched matmul
- Uses cublasGemmBatchedEx introduced in CUDA 9.1.
- Includes support for Tensor Op math.
- Falls back to a loop over non-batched gemm calls on older CUDA
versions or GPU architectures.
* Refactor GPU batched gemm into one internal func
andrehentz [Thu, 10 May 2018 17:55:23 +0000 (10:55 -0700)]
Merge pull request #17927 from smitshilu/patch-1
Update tensorflow/contrib/lite/BUILD
Amit Patankar [Thu, 10 May 2018 17:12:32 +0000 (10:12 -0700)]
Merge pull request #19197 from smitshilu/patch-4
Updated index.md
Sergio Guadarrama [Thu, 10 May 2018 16:08:43 +0000 (09:08 -0700)]
Add citation (#19201)
Yong Tang [Thu, 10 May 2018 16:07:44 +0000 (09:07 -0700)]
Fix warning in python 3 with deprecated inspect.getargspec (#19199)
This fix tries to address the issue raised in 16152 where
a warning will show up in python 3 with:
```
import tensorflow as tf
import warnings
warnings.filterwarnings('error')
tf.reduce_sum(tf.placeholder(tf.float64))
......
DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() or inspect.getfullargspec()
```
This fixes the issue with getfullargspec in tf_export, which takes
into consideration the python 2 vs python3.
This fix fixes 16152.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Smit Shilu [Thu, 10 May 2018 13:19:31 +0000 (09:19 -0400)]
Updated index.md
Tensorflow lite image was not visible. Added image file path
AG Ramesh [Wed, 9 May 2018 16:51:10 +0000 (09:51 -0700)]
[INTEL MKL] Fixes a failure in //tensorflow/python/profiler:model_analyzer_test. (#19152)
* Modified testComplexCodeView test
Modified testComplexCodeView to look for lower total_float_ops. The value of total_float_ops is lower when Tensorflow is compiled with Intel MKL.
* Added code to check if MKL is enabled
* Fixed Pylint errors
Letian Feng [Wed, 9 May 2018 16:50:57 +0000 (18:50 +0200)]
Fix 2 typos in documents (#19177)
* fix minor typo in doc: tf.layer to tf.layers
* removed a duplicated line
ctiijima [Wed, 9 May 2018 16:50:48 +0000 (09:50 -0700)]
Grammar fixes for Programmers guide FAQ (#19170)
Yong Tang [Wed, 9 May 2018 15:55:19 +0000 (08:55 -0700)]
Enable test case for float64 with conv1d (#19179)
The float64 for conv2d support has been added to tensorflow
in
e3468b56d323783fdfb79fa2d6c24effc58bcaa9. (Thanks brianwa84!)
Since conv1d implementation invokes conv2d, the float64 support
for conv1d is supported now as well.
This fix adds the test case for float64 support of conv1d and
removes the TODO.
This fix fixes 19175.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
Yifei Feng [Wed, 9 May 2018 04:59:43 +0000 (21:59 -0700)]
Merge pull request #19164 from yifeif/branch_195897321
Branch
195897321
Yifei Feng [Wed, 9 May 2018 01:24:40 +0000 (18:24 -0700)]
Merge pull request #19165 from tensorflow/update_version_fix
Switch to use str instead of number for colab_url
Yifei Feng [Wed, 9 May 2018 00:35:21 +0000 (17:35 -0700)]
Switch to use str instead of number for colab_url
Fix nightly failure:
File "tensorflow/tools/ci_build/update_version.py", line 253, in colab_url
version_string = "%d.%d.%d" % (version.major, version.minor, version.patch)
TypeError: %d format: a number is required, not str
Yifei Feng [Wed, 9 May 2018 00:16:45 +0000 (17:16 -0700)]
Hardcode EndpointSpec deprecated input to False for now after cl/
195718061.
Yifei Feng [Wed, 9 May 2018 00:14:55 +0000 (17:14 -0700)]
Merge commit for internal changes
Igor Ganichev [Tue, 8 May 2018 23:43:54 +0000 (16:43 -0700)]
Make eager functions runable on TPU
PiperOrigin-RevId:
195897321
A. Unique TensorFlower [Tue, 8 May 2018 23:23:27 +0000 (16:23 -0700)]
Set size of tensorflow/python/keras:normalization_test to medium to avoid flaky timeouts
PiperOrigin-RevId:
195894737
A. Unique TensorFlower [Tue, 8 May 2018 23:20:02 +0000 (16:20 -0700)]
Only use integer values for event_ndims.
event_ndims have the semantics of being an integer. However, other code paths (such as const_value)
can return back numpy wrapped arrays, which can mess with how values are cached. Instead extract
everything as an integer.
PiperOrigin-RevId:
195894216
A. Unique TensorFlower [Tue, 8 May 2018 23:16:57 +0000 (16:16 -0700)]
add test for pruning useless function lib in graph.
PiperOrigin-RevId:
195893756
A. Unique TensorFlower [Tue, 8 May 2018 22:43:34 +0000 (15:43 -0700)]
Add test to test suite.
PiperOrigin-RevId:
195888932
A. Unique TensorFlower [Tue, 8 May 2018 22:34:40 +0000 (15:34 -0700)]
Increase shard count of tensorflow/contrib/learn:state_saving_rnn_estimator_test to avoid flaky timeouts
PiperOrigin-RevId:
195887546
A. Unique TensorFlower [Tue, 8 May 2018 22:33:37 +0000 (15:33 -0700)]
Increase size of tensorflow/contrib/distributions:batch_reshape_test to medium to avoid flaky timeouts
PiperOrigin-RevId:
195887374
Lukas Geiger [Tue, 8 May 2018 23:31:39 +0000 (01:31 +0200)]
[tfgan] Allow to add custom eval metrics to GANEstimator (#19133)
jjsjann123 [Tue, 8 May 2018 23:30:08 +0000 (16:30 -0700)]
[tftrt update] (#19135)
* [tftrt update]
code cleaning, removed some boilerplate code
* addressing comments
A. Unique TensorFlower [Tue, 8 May 2018 22:26:44 +0000 (15:26 -0700)]
Increase shard count of tensorflow/python/keras:lstm_test to avoid flaky timeouts
PiperOrigin-RevId:
195886372
A. Unique TensorFlower [Tue, 8 May 2018 21:50:24 +0000 (14:50 -0700)]
Add missing ":haswell" match to list of platform selectors.
PiperOrigin-RevId:
195880275
A. Unique TensorFlower [Tue, 8 May 2018 21:45:01 +0000 (14:45 -0700)]
Increase shard_count of tensorflow/python/estimator:estimator_test to avoid flaky asan timeouts
PiperOrigin-RevId:
195879364
Alexandre Passos [Tue, 8 May 2018 21:42:35 +0000 (14:42 -0700)]
Do not differentiage integers in the eager API.
This is similar to the change made in:
https://github.com/tensorflow/tensorflow/commit/
f63750645826df65b05cad505546a86f0e347674
for backpropagation during graph construction via tf.gradients()
PiperOrigin-RevId:
195878952
A. Unique TensorFlower [Tue, 8 May 2018 21:41:48 +0000 (14:41 -0700)]
Increase shard count of tensorflow/contrib/distributions:mixture_test to avoid flaky timeouts in asan mode
PiperOrigin-RevId:
195878809
A. Unique TensorFlower [Tue, 8 May 2018 21:00:48 +0000 (14:00 -0700)]
Increase size of test tensorflow/contrib/layers:rev_block_lib_test to medium
to avoid flaky timeouts.
PiperOrigin-RevId:
195871947
Allen Lavoie [Tue, 8 May 2018 21:00:30 +0000 (14:00 -0700)]
Avoid string formatting in assert_same_float_dtype unless there's an error
Especially helpful when executing eagerly
PiperOrigin-RevId:
195871887
A. Unique TensorFlower [Tue, 8 May 2018 19:10:36 +0000 (12:10 -0700)]
Better wrapping of stream executor's cuDNN API calls. Replacing mutex locking and setting the cuDNN stream followed by calling wrap::cudnn... with an RAII CudnnHandle object that handles the former two operations.
Distinguish three different API types:
A) APIs that don't take a cudnnHandle_t: These are thread-safe APIs that don't enqueue any CUDA work on a stream. They can be called directly without any extra precautions.
B) APIs that take a cudnnHandle_t and perform CUDA work. The CUDA context needs to be acquired and the stream needs to be set beforehand, calls need to be serialized. A CudnnHandle instance guarantees that this work has been performed before calling cuDNN.
C) APIs that do take a cudnnHandle_t, but (presumably, the API makes no guarantees) still don't perform any CUDA work. This is limited to the API to setup RNN descriptors. Calls need to be serialized, but most likely we wouldn't need to acquire the CUDA context or set the stream. We still do though using the legacy default stream, because there are no guarantees.
PiperOrigin-RevId:
195856300
A. Unique TensorFlower [Tue, 8 May 2018 19:04:38 +0000 (12:04 -0700)]
Remove outdated CUDA SDK string (the text is now consistent with other version choices, and the '9.0' format is already present in the default).
PiperOrigin-RevId:
195855416
Alina Sbirlea [Tue, 8 May 2018 18:54:03 +0000 (11:54 -0700)]
Re-land: Optimize dot(DynamicSlice(ConstA), ConstantB) by memoizing dot(ConstA, ConstB)
Make transformation when ConstA and ConstB are 2D, and DynamicSlice is slicing a full row, column respectively.
Handle:
dot(DynamicSlice(Index, ConstA), ConstB) => DynamicSlice(Index, dot*(ConstA, ConstB));
and
dot(ConstA, DynamicSlice(Index, ConstB)) => DynamicSlice(Index, dot*(ConstA, ConstB));
Reason to roll forward: Previous issue of getting out of memory errors when generating LLVM constants was resolved by CSE-ing constants before allocation.
PiperOrigin-RevId:
195853680
A. Unique TensorFlower [Tue, 8 May 2018 18:45:53 +0000 (11:45 -0700)]
Fix docstring for flush() method
PiperOrigin-RevId:
195852402
Ilya Biryukov [Tue, 8 May 2018 18:25:50 +0000 (11:25 -0700)]
Update version of downloadable clang toolchain
PiperOrigin-RevId:
195849091
A. Unique TensorFlower [Tue, 8 May 2018 18:19:46 +0000 (11:19 -0700)]
Change visibility of hlo_proto.
PiperOrigin-RevId:
195848035
A. Unique TensorFlower [Tue, 8 May 2018 18:15:53 +0000 (11:15 -0700)]
Add affinity binding functionality and documentation to OVIC benchmarker.
PiperOrigin-RevId:
195847378
A. Unique TensorFlower [Tue, 8 May 2018 18:12:07 +0000 (11:12 -0700)]
Increase size of test //third_party/tensorflow/python:saver_large_variable_test
from "small" to "medium" to prevent flaky timeouts.
PiperOrigin-RevId:
195846802
Andrew Selle [Tue, 8 May 2018 18:10:23 +0000 (11:10 -0700)]
Fix Raspberry Pi build by making PNG not try to use Neon (by autodetect).
This involves patching to override the png neon option. In the future
it might be worth enabling PNG optimization.
PiperOrigin-RevId:
195846513
Akshay Agrawal [Tue, 8 May 2018 18:07:45 +0000 (11:07 -0700)]
When building functions, capture tensors in `internal_convert_to_tensor`.
This change is motivated by the fact that, when eager execution is disabled, library functions assume that tensors returned from `internal_convert_to_tensor` are in fact `Tensor`s and not `EagerTensor`s.
PiperOrigin-RevId:
195846039
A. Unique TensorFlower [Tue, 8 May 2018 17:25:30 +0000 (10:25 -0700)]
Add cost model of depthwiseConv2dNative. Tensorflow computes depthwise separable convolutions as depthwiseConv2dNative followed by 1x1 Conv2D
PiperOrigin-RevId:
195838887
A. Unique TensorFlower [Tue, 8 May 2018 16:46:45 +0000 (09:46 -0700)]
Free ANeuralNetworksCompilation object in NNAPIDelegate destructor
PiperOrigin-RevId:
195832807
Shanqing Cai [Tue, 8 May 2018 16:04:17 +0000 (09:04 -0700)]
Minor formatting tweaks to distribute.py and simple_tfkeras_example.py
PiperOrigin-RevId:
195827029
A. Unique TensorFlower [Tue, 8 May 2018 15:57:45 +0000 (08:57 -0700)]
Update comment clarifying continuous eval behavior.
PiperOrigin-RevId:
195826025
Peter Hawkins [Tue, 8 May 2018 15:07:08 +0000 (08:07 -0700)]
[TF:XLA] Fix NaN in StatelessRandomNormal if the underlying uniform distribution returned -1.
PiperOrigin-RevId:
195819645
A. Unique TensorFlower [Tue, 8 May 2018 15:04:07 +0000 (08:04 -0700)]
Automated g4 rollback of changelist
195723288
PiperOrigin-RevId:
195819297
A. Unique TensorFlower [Tue, 8 May 2018 14:57:12 +0000 (07:57 -0700)]
Add missing #include for OpResponse. This class currently happens to be forward
declared by xla.proto.h, but that proto doesn't actually need this type
anywhere and we are working on removing such unneeded forward declarations.
PiperOrigin-RevId:
195818397
Asim Shankar [Tue, 8 May 2018 14:28:43 +0000 (07:28 -0700)]
ProfileHandler: Remove unnecessary interface method.
PiperOrigin-RevId:
195815565
A. Unique TensorFlower [Tue, 8 May 2018 10:26:06 +0000 (03:26 -0700)]
Fix a test expectation.
PiperOrigin-RevId:
195796348
A. Unique TensorFlower [Tue, 8 May 2018 09:11:52 +0000 (02:11 -0700)]
Automated g4 rollback of changelist
195748721
PiperOrigin-RevId:
195790581
A. Unique TensorFlower [Tue, 8 May 2018 02:56:26 +0000 (19:56 -0700)]
Temporarily disable concat rewrite.
PiperOrigin-RevId:
195762860
Brennan Saeta [Tue, 8 May 2018 01:31:47 +0000 (18:31 -0700)]
[tf.data] Move tensorflow::dataset::MakeIteratorContext to core/framework
PiperOrigin-RevId:
195756342
Martin Wicke [Tue, 8 May 2018 19:38:56 +0000 (12:38 -0700)]
Merge pull request #19053 from wangsiyu/fix_tf_layers_reuse
Fix bug of declaring regularization loss multiple times when reusing PartitionedVariables in tf layers
wangsiyu [Tue, 8 May 2018 07:50:36 +0000 (15:50 +0800)]
remove type
wangsiyu [Tue, 8 May 2018 07:46:38 +0000 (15:46 +0800)]
fix bug of return value in _should_add_regularizer function and refine code in base_test.py to make it no more than 80 columns
wangsiyu [Tue, 8 May 2018 04:58:27 +0000 (12:58 +0800)]
eliminate result variable in _should_add_regularizer to make code clean
wangsiyu [Tue, 8 May 2018 02:54:04 +0000 (10:54 +0800)]
refine unit test case coding style and move _should_add_regularizer function into add_weight
Skye Wanderman-Milne [Tue, 8 May 2018 00:28:41 +0000 (17:28 -0700)]
Add logic for StridedSlice ops in ShapeRefiner::ConstantPartialShape().
This mimics the logic in tensor_util.constant_value_as_shape, allowing
the C++ shape inference code to infer more shapes than it could before.
This change also adds an optional stride argument to InferenceContext::Subshape().
PiperOrigin-RevId:
195749522
Skye Wanderman-Milne [Tue, 8 May 2018 00:24:28 +0000 (17:24 -0700)]
Make conv2d_tranpose_test.py work with C API shapes enabled.
The C API provides more accurate shape information in many cases.
PiperOrigin-RevId:
195749030
Igor Ganichev [Tue, 8 May 2018 00:21:53 +0000 (17:21 -0700)]
Make eager functions runable on TPU
PiperOrigin-RevId:
195748721
Skye Wanderman-Milne [Tue, 8 May 2018 00:21:39 +0000 (17:21 -0700)]
Raise an error if we try to take the gradient wrt to the initial value of a loop variable.
Fixes #14101
PiperOrigin-RevId:
195748688
Blake Hechtman [Tue, 8 May 2018 00:00:27 +0000 (17:00 -0700)]
Internal change
PiperOrigin-RevId:
195745819
Jacques Pienaar [Mon, 7 May 2018 23:59:41 +0000 (16:59 -0700)]
Add test with tf.cond.
PiperOrigin-RevId:
195745718
Sanjoy Das [Mon, 7 May 2018 23:55:10 +0000 (16:55 -0700)]
Delete kTransposeDot (it is no longer in use)
PiperOrigin-RevId:
195745124
Alexandre Passos [Mon, 7 May 2018 23:49:44 +0000 (16:49 -0700)]
Fast-path to VarHandleOp
PiperOrigin-RevId:
195744374
Billy Lamberta [Mon, 7 May 2018 23:38:02 +0000 (16:38 -0700)]
Add TFX section. Add Ecosystem page and dropdown menu.
PiperOrigin-RevId:
195742728
A. Unique TensorFlower [Mon, 7 May 2018 23:31:07 +0000 (16:31 -0700)]
Reorder executor NodeItem variable length data section so
that all multi-byte aligned types precede all byte-aligned
types so that alignment is satisfied without padding.
PiperOrigin-RevId:
195741712
Skye Wanderman-Milne [Mon, 7 May 2018 23:16:32 +0000 (16:16 -0700)]
ShapeRefiner fix: some variant-type tensors have handle data.
ShapeRefiner::AddNode() would only propagate handle data for
DT_RESOURCE tensors, but not DT_VARIANT. The Python shape inference
logic in common_shapes.py handled this correct, which is why we didn't
notice this earlier. In particular, list ops use DT_VARIANT with
handle data.
PiperOrigin-RevId:
195739586
Akshay Agrawal [Mon, 7 May 2018 23:16:24 +0000 (16:16 -0700)]
Refactor TensorArray to avoid copies and memory allocations when executing eagerly.
With this change, writes to TensorArrays when eager execution is enabled take O(1) time instead of O(n). Additionally, whereas writing to a TensorArray when constructing a graph results in allocating a new Python TensorArray object, writing to a TensorArray with eager enabled no longer performs that allocation (graph construction uses these allocations to ensure correctness of control flow and gradients, but this isn't necessary when executing eagerly). Finally, this change also removes the artificial write-once semantics of TensorArrays when executing eagerly.
PiperOrigin-RevId:
195739572
Blake Hechtman [Mon, 7 May 2018 22:58:29 +0000 (15:58 -0700)]
[XLA] Make post order a possible schedule as it sometimes uses less memory than
the DFS or list scheduler and it is very simple.
PiperOrigin-RevId:
195736916
Derek Murray [Mon, 7 May 2018 22:51:05 +0000 (15:51 -0700)]
[Remote functions] Only set the default runner *after* resolving the remote FLR.
Previously, if the `runner` was not specified for a function
execution, we would immediately set it to the default runner of the
*local* FLR, even if the function was to be executed remotely. This
change postpones the resolution of the default runner until after the
function invocation has been routed to the FLR that will actually
execute it.
As a result, we avoid the pathological case where a GPU device using a
private threadpool (TF_GPU_THREAD_MODE=gpu_private) ends up running
all of the ops for the CPU-side input pipeline on the private
threadpool.
PiperOrigin-RevId:
195735734
A. Unique TensorFlower [Mon, 7 May 2018 22:47:57 +0000 (15:47 -0700)]
Add EvaluateNodes to tests: RemoveIdentityTransposesMultipleOutputs, RemoveTransposesWithControlDependency, CombineBitcasts, CombineAndRemoveBitcasts, RemoveRedundantCast
PiperOrigin-RevId:
195735234
Michael Kuperstein [Mon, 7 May 2018 22:41:52 +0000 (15:41 -0700)]
[XLA] Fix a "we're we're" in the operation semantics.
PiperOrigin-RevId:
195734316
A. Unique TensorFlower [Mon, 7 May 2018 22:41:22 +0000 (15:41 -0700)]
Add support for select (via tf.where) support to tflite.
PiperOrigin-RevId:
195734246
Michael Case [Mon, 7 May 2018 22:24:02 +0000 (15:24 -0700)]
Internal Change.
PiperOrigin-RevId:
195731675
Michael Case [Mon, 7 May 2018 22:20:49 +0000 (15:20 -0700)]
Fix TypeError in update_version.py
PiperOrigin-RevId:
195731183
Pavithra Vijay [Mon, 7 May 2018 22:16:59 +0000 (15:16 -0700)]
Add support for tf.data.Dataset iterators in model training/eval methods in eager-mode
PiperOrigin-RevId:
195730534
Igor Ganichev [Mon, 7 May 2018 22:15:01 +0000 (15:15 -0700)]
Replace references to TensorInfo with XlaTensor
PiperOrigin-RevId:
195730139