Gitiles
Code Review
Sign In
gerrit-public.fairphone.software
/
platform
/
external
/
tensorflow
/
a997e989c5559d42ddbc253df09e5e8b13557dd1
a997e98
[TF:XLA] Always return a tuple-shaped result when converting a TensorFlow graph into an XLA computation. Previously we had special case logic where the number of outputs was 1 to avoid the tuple, but this meant that any code that wanted to reason about the outputs had to have a special case. By using tuples unconditionally we can simplify the code.
by Peter Hawkins
· 8 years ago
36cf543
Enable GPU support of QR decomposition for the cases left out in cl/169249336.
by A. Unique TensorFlower
· 8 years ago
563c2a4
Adds continuous evaluation (for distributed train and eval) into _TrainingExecutor.
by Jianwei Xie
· 8 years ago
e103bae
Merge the implementations of XlaLocalLaunch and XlaDeviceLaunch.
by A. Unique TensorFlower
· 8 years ago
ada8d9d
internal
by A. Unique TensorFlower
· 8 years ago
ff95455
Add a platform-dependent self-check that runs during `import tensorflow`.
by Derek Murray
· 8 years ago
76293c2
Fixes for bare-column and column generators as inputs
by Mark Daoust
· 8 years ago
d17e073
Added a check to ensure # of queue components match the list of tensors getting
by A. Unique TensorFlower
· 8 years ago
7351390
Add an env-var TF_ENABLE_FFT_TILING_FORWARD.
by Yangzihao Wang
· 8 years ago
d67679f
Implement TF_FunctionImportFunctionDef
by Igor Ganichev
· 8 years ago
74680a3
Internal change.
by Anna R
· 8 years ago
23ac8df
Multi_head for core Estimators.
by A. Unique TensorFlower
· 8 years ago
3578491
[XLA] Add support for convolution to ShapeVerifier
by David Majnemer
· 8 years ago
b211be1
TFE: Simplify code example in backprop.py
by Shanqing Cai
· 8 years ago
2bf4abe
Roll foward of changing global step to int64.
by Jacques Pienaar
· 8 years ago
0905a17
eager: Export tfe.Iterator for iterating over tf.contrib.data.Datasets
by Asim Shankar
· 8 years ago
4d53bb2
Pass around the eager context object instead of getting it
by A. Unique TensorFlower
· 8 years ago
f78a2ee
Bugfix: Correct batch-shape broadcast bugs in tf.contrib.distributions.VectorDiffeomixture.
by Joshua V. Dillon
· 8 years ago
4162c82
Apply const qualifiers whenever appropriate.
by Henry Tan
· 8 years ago
8c56120
Changing the default pipeline for CNN benchmark to use datasets.
by Jiri Simsa
· 8 years ago
d52cff9
ApiDef will replace OpGenOverrides proto and take over responsibility for
by Anna R
· 8 years ago
0666fa2
Parameterize tests for select-and-scatter
by Yunxing Dai
· 8 years ago
10155f7
Improvements in graph_properties.cc:
by Max Galkin
· 8 years ago
52871a8
Adding missing deps to targets in llvm.BUILD. This was only working in non-sandboxed builds.
by A. Unique TensorFlower
· 8 years ago
6e6f767
Fixed a flaky test.
by Mustafa Ispir
· 8 years ago
4a02793
PUBLIC: [tf.contrib.data] Standardize transformation functions for use with `Dataset.apply()`.
by Derek Murray
· 8 years ago
7f56f6c
Adding a copy binary script to GitHub to change python versions for whl files.
by Amit Patankar
· 8 years ago
a249949
Fix build errors caused by #12502
by A. Unique TensorFlower
· 8 years ago
8a1fb64
Adding a copy binary script to GitHub to change python versions for whl files.
by Amit Patankar
· 8 years ago
20f7eb3
Fix build errors caused by #12502
by A. Unique TensorFlower
· 8 years ago
e9bd04a
Supports label_vocabulary in multi_label_head.
by A. Unique TensorFlower
· 8 years ago
3588ff7
Migrate `instance_norm` to contrib/layers.
by A. Unique TensorFlower
· 8 years ago
5b94356
Fix typo in convert_image_to_dtype
by Allen Lavoie
· 8 years ago
095f6aa
Fix ops namespace to boosted_trees.
by A. Unique TensorFlower
· 8 years ago
34018f8
Add GPU support for QR decomposition.
by A. Unique TensorFlower
· 8 years ago
ec962ff
Support specifying output names during TF_Function creation
by Igor Ganichev
· 8 years ago
e4223cb
Generate dummy crosstool_wrapper_driver_is_not_gcc with cuda clang configuration so that it is possible to build bazel build @local_config_cuda//...
by A. Unique TensorFlower
· 8 years ago
5882ae3
Update to NCCL version 1.3.5. Remove temporary buffer for ncclReduce, it's no longer needed in this version.
by A. Unique TensorFlower
· 8 years ago
23da211
Add liveness_util functions which use dataflow analysis. Also make the analysis argument (TuplePointsToAnalysis or HloDataflowAnalysis) non-optional as all callers were passing in the analysis.
by Mark Heffernan
· 8 years ago
f08ec57
TF Eager: Avoid creating some unnecessary zeros during backprop.
by A. Unique TensorFlower
· 8 years ago
dba4bf1
eager: Add some more top-level API tests.
by Asim Shankar
· 8 years ago
1da763a
Add function gradient support to C API
by Igor Ganichev
· 8 years ago
d10902f
[XLA] Fix HLO-graph-dump handling of nodes near computation edges, and indicate root node.
by A. Unique TensorFlower
· 8 years ago
d991ac0
Internal change.
by Jianwei Xie
· 8 years ago
ce78e8d
Cut an additional dependency from xla:service on the GPU backend.
by Justin Lebar
· 8 years ago
0429f1c
[TF:XLA] Simplify XlaCompiler API by moving XlaCompiler::BuildExecutable() into XlaCompilationCache, which is its only caller. No functional changes.
by Peter Hawkins
· 8 years ago
15d2b7a
[TF:XLA] calls tf.TensorFlowTestCase's setUp/tearDown in the setUp/tearDown
by A. Unique TensorFlower
· 8 years ago
e5ed09b
Fix docs parser to handle keras docs
by Mark Daoust
· 8 years ago
f0550d0
Include stride information for slice in hlo graph dump and hlo string.
by A. Unique TensorFlower
· 8 years ago
04eb945
Protect XLA_{V,}LOG_LINES with a mutex.
by Justin Lebar
· 8 years ago
662a2e6
Update the checkpoints index file in CheckpointProto before actually deleting files.
by Igor Saprykin
· 8 years ago
7de939b
[TF:XLA] Create Interpreter backend from the Executor backend.
by Kay Zhu
· 8 years ago
de724b1
Add anonymous namespace in compiler_functor.cc.
by Justin Lebar
· 8 years ago
20d2fa0
Store xla_metadata as a private field of XlaDevice.
by Jianwei Xie
· 8 years ago
5378259
Create empty directory for upcoming K-FAC library.
by A. Unique TensorFlower
· 8 years ago
12933c1
Annotate more loops in LLVM IR with their HLO names.
by Justin Lebar
· 8 years ago
9dc8947
Rolling forward a reduction in the number of strided slice constants
by Allen Lavoie
· 8 years ago
4727531
TFE: Add default value back to params of value_and_gradients_function
by Shanqing Cai
· 8 years ago
225ae5a
Internal changes
by Shanqing Cai
· 8 years ago
65866f0
TFE: Add Python code examples in doc strings and make existing ones executable
by Shanqing Cai
· 8 years ago
6081480
PiperOrigin-RevId: 169059046
by A. Unique TensorFlower
· 8 years ago
65fff34
Add TFGAN README.
by A. Unique TensorFlower
· 8 years ago
381357f
Create PyRaiseException() with common code to convert from a C string
by A. Unique TensorFlower
· 8 years ago
fe06f11
Fix capitalization of some TFE/Python C APIs.
by A. Unique TensorFlower
· 8 years ago
f46f3f3
Cleaner exception traces in Python 3 when eager execution triggers an error.
by A. Unique TensorFlower
· 8 years ago
62b8a74
Move Python sequence -> TensorFlow Tensor conversion from Python/NumPy to C/C++.
by A. Unique TensorFlower
· 8 years ago
700711d
Update LLVM version. Follow LLVM API change.
by A. Unique TensorFlower
· 8 years ago
c606b0c
Move Python sequence -> TensorFlow Tensor conversion from Python/NumPy to C/C++.
by A. Unique TensorFlower
· 8 years ago
3aedb9b
Fix the profiler copy.bara
by A. Unique TensorFlower
· 8 years ago
20e12d9
Enabling sparsify_gather to be called before freeze.
by A. Unique TensorFlower
· 8 years ago
c2749f9
Fixed formatting in contrib.data.Dataset documentation.
by A. Unique TensorFlower
· 8 years ago
2abb21c
Use the correct module for ones_like.
by A. Unique TensorFlower
· 8 years ago
827fe7b
[XLA:GPU] Use CallInliner pass to work around broken Call support on GPU. Enable tests that use Call.
by Peter Hawkins
· 8 years ago
1d04d82
Add ability to load graph from resource.
by A. Unique TensorFlower
· 8 years ago
a7e62ba
Cut dependencies between XLA client/service and CPU/GPU backends.
by Justin Lebar
· 8 years ago
b5e080d
Remove contrib/imperative.
by Vincent Vanhoucke
· 8 years ago
6697af4
Dynamically control training/evaluation iterations.
by Jianwei Xie
· 8 years ago
e52036a
Go: Update generated wrapper functions for TensorFlow ops.
by A. Unique TensorFlower
· 8 years ago
15ea74a
Update ops-related pbtxt files.
by A. Unique TensorFlower
· 8 years ago
a373b1f
Merge changes from github. END_PUBLIC
by Patrick Nguyen
· 8 years ago
c48d696
[XLA:CPU] Implement map fusion.
by Sanjoy Das
· 8 years ago
87f5182
Fix broken test: tensorflow/contrib/eager/python:datasets_test
by Asim Shankar
· 8 years ago
bd3ba66
Internal change.
by Eugene Brevdo
· 8 years ago
154e4e5
disable asan for test timeout
by Olivia Nordquist
· 8 years ago
4ad32ed
disabling failing tsan test
by Olivia Nordquist
· 8 years ago
b57bbde
Adding support for capture-by-value.
by A. Unique TensorFlower
· 8 years ago
2dcdd6c
Update the minimum requirement of espsilon for batch norm.
by Yao Zhang
· 8 years ago
bc68dc8
Add ops that perform color transforms (including changing value, saturation and hue) in YIQ space.
by A. Unique TensorFlower
· 8 years ago
30868ef
disabling timeout test in opensource build
by Olivia Nordquist
· 8 years ago
0f17dab
[tf.contrib.data] Expose `tf.contrib.data.batch_and_drop_remainder()`.
by Derek Murray
· 8 years ago
e8d40d1
Adds DNNEstimator to tf.contrib.estimator.
by A. Unique TensorFlower
· 8 years ago
735e7e2
Automated g4 rollback of changelist 168458634
by Patrick Nguyen
· 8 years ago
439a2b7
Output metrics in train mode for multihead.
by A. Unique TensorFlower
· 8 years ago
f9ab7c9
Make sure all checked threads are joined before they are terminated.
by Gunhan Gulsoy
· 8 years ago
8b96630
Fixing the build command.
by A. Unique TensorFlower
· 8 years ago
6c5bc8c
Script that generates __init__.py files based on tf_api_names annotations.
by Anna R
· 8 years ago
2691325
Adds multi_label_head in tf.contrib.estimator
by A. Unique TensorFlower
· 8 years ago
75dbd12
tfdbg: fix a bug in list_inputs and list_outputs
by Shanqing Cai
· 8 years ago
0be0c11
[XLA] Add S64 while loop test.
by Jacques Pienaar
· 8 years ago
b4224fe
Certain ops don't need eager gradients to keep their inputs / outputs alive.
by Alexandre Passos
· 8 years ago
Next »