aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Disable tests that are currently failing with cuda 9Gravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191805453
* [XLA] Better support for mul reductions in MakeFakeArguments()Gravatar Michael Kuperstein2018-04-05
| | | | | | Mul reductions want a 1 as their init value, not a 0 or a random value. PiperOrigin-RevId: 191802819
* Fix TF_ImportGraphDefResults and TF_Function leaks in Python API.Gravatar Skye Wanderman-Milne2018-04-05
| | | | PiperOrigin-RevId: 191797853
* quantized LSTM support improvementsGravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191794956
* Changes loss_reduction default to SUM_OVER_BATCH_SIZE for multi_class_head ↵Gravatar A. Unique TensorFlower2018-04-05
| | | | | | and binary_classification_head. PiperOrigin-RevId: 191793392
* Enable branch prediction in TensorFlowGravatar Justine Tunney2018-04-05
| | | | PiperOrigin-RevId: 191788253
* [XLA] Remove a dead function and a stale todo.Gravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191786563
* Automated g4 rollback of changelist 191753026Gravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191784709
* Save the original from_proto method before calling it to avoid infinite loop.Gravatar Priya Gupta2018-04-05
| | | | PiperOrigin-RevId: 191784430
* Fix regression caused by cl/191020868: Re-use materialized shapes for other ↵Gravatar A. Unique TensorFlower2018-04-05
| | | | | | broadcast gradient shape nodes. PiperOrigin-RevId: 191779263
* Fix final eval bottleneck creation to work in cases where it isn't cached ↵Gravatar Suharsh Sivakumar2018-04-05
| | | | | | | | already. Fixes #17423 PiperOrigin-RevId: 191773001
* Automated g4 rollback of changelist 191761109Gravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191771969
* Expose odeint_fixed in tf.contrib.integrateGravatar Stephan Hoyer2018-04-05
| | | | PiperOrigin-RevId: 191769890
* Internal changeGravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191769724
* Sync only the convolutional_recurrent file to Keras 2.1.5.Gravatar Anjali Sridhar2018-04-05
| | | | PiperOrigin-RevId: 191763101
* Inline more functionsGravatar Benoit Steiner2018-04-05
| | | | PiperOrigin-RevId: 191761109
* Expand activity analysis to the test nodes of if and while statements.Gravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191756234
* Include the operators module in the test framework as well.Gravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191756100
* Refine BatchReshape error messages.Gravatar Joshua V. Dillon2018-04-05
| | | | PiperOrigin-RevId: 191754120
* Document expected regular structure of the statistical testing library.Gravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191753693
* Added a call in CheckpointSaverHook.after_create_session to always saveGravatar Sherry Moore2018-04-05
| | | | | | checkpoint before the first training step. PiperOrigin-RevId: 191753026
* Fix typos in "Profile Model Float Operations" documentation.Gravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191751175
* Use constants in tf.zeros if the constant won't be too big.Gravatar Skye Wanderman-Milne2018-04-05
| | | | | | | | Using fill saves on GraphDef size, but can slow down models since the total number of ops is greater (fill + shape + constant op). This change makes us only use fill for large shapes. PiperOrigin-RevId: 191747456
* Fix docstring.Gravatar A. Unique TensorFlower2018-04-05
| | | | PiperOrigin-RevId: 191747417
* Create a separate operators module that is to contain all Python constructs ↵Gravatar A. Unique TensorFlower2018-04-05
| | | | | | that we override: control flow, builtins, operators, etc. PiperOrigin-RevId: 191729654
* Add support for NCCL2. The configure script asks for what version of NCCL to ↵Gravatar A. Unique TensorFlower2018-04-05
| | | | | | | | use. The default is still NCCL 1 from GitHub. If the user chooses NCCL 2, it asks for the install directory. The nccl_configure.bzl generates two different BUILD files based on the chose NCCL version. For NCCL 1, it aliases to the existing 'nccl_archive' http_repo on GitHub. For NCCL 2, it creates a target containing the NCCL 2 library and headers from the chosen install directory. PiperOrigin-RevId: 191718007
* Upgrade Snappy to latest versionGravatar Justine Tunney2018-04-05
| | | | | | | They added big endian preprocessor macros in recent releases. Hopefully this should do the right thing on IBM mainframes. PiperOrigin-RevId: 191705207
* Upgrade zlib to latest versionGravatar Justine Tunney2018-04-04
| | | | PiperOrigin-RevId: 191696213
* Support arbitrary permutations for DataFormatDimMap.Gravatar A. Unique TensorFlower2018-04-04
| | | | PiperOrigin-RevId: 191696203
* Add a helper function to re-assign colocation in a graph.Gravatar A. Unique TensorFlower2018-04-04
| | | | PiperOrigin-RevId: 191679495
* Sort control inputs alphabetically in ToGraphDefSubRange.Gravatar Skye Wanderman-Milne2018-04-04
| | | | PiperOrigin-RevId: 191677358
* BUGFIX: Detect when broadcasting is required and raise NotImplementedError.Gravatar Joshua V. Dillon2018-04-04
| | | | PiperOrigin-RevId: 191673876
* Exclude some gcc options in Windows buildGravatar A. Unique TensorFlower2018-04-04
| | | | PiperOrigin-RevId: 191672761
* [XLA]: Enable Memory sanitizer for compilation_passes_test.Gravatar Bixia Zheng2018-04-04
| | | | | | | Initialize the buffer for a scalar tensor to avoid uninitialized accesses. PiperOrigin-RevId: 191672257
* Replace trivial backend calls with calls to underlying TensorFlow functions ↵Gravatar Pavithra Vijay2018-04-04
| | | | | | - Part 2 PiperOrigin-RevId: 191669725
* Address bug in distributed strategies `Monitor` to allow running for >1 step.Gravatar Priya Gupta2018-04-04
| | | | PiperOrigin-RevId: 191667378
* Check that n + kBlockTrailerSize does not overflow before reading a blockGravatar Frank Chen2018-04-04
| | | | PiperOrigin-RevId: 191666300
* Fixed handling of control dependencies in the arithmethic optimizerGravatar Benoit Steiner2018-04-04
| | | | PiperOrigin-RevId: 191665098
* Running sparse_ops_test only in opt mode since the test is flaky (times out) ↵Gravatar A. Unique TensorFlower2018-04-04
| | | | | | in fastbuild mode. PiperOrigin-RevId: 191665014
* Enable constant propagation across Enter nodes, but only if is_constant is true.Gravatar A. Unique TensorFlower2018-04-04
| | | | | | Don't propagate constants with control dependencies through Merge nodes. PiperOrigin-RevId: 191663396
* Internal change.Gravatar A. Unique TensorFlower2018-04-04
| | | | PiperOrigin-RevId: 191660588
* Compile TensorFlow with /arch:AVX on WindowsGravatar A. Unique TensorFlower2018-04-04
| | | | | | /arch:AVX is the correpsonding option in MSVC for gcc's --march=native PiperOrigin-RevId: 191660389
* Adds commandline option (toco_compatible, bool) that makes the ↵Gravatar A. Unique TensorFlower2018-04-04
| | | | | | | | optimize_for_inference script only use TOCO friendly ops. In particular, FusedResizeAndPadConv2D is not supported by TOCO. This change does not alter existing behavior (the boolean is set to false by default). PiperOrigin-RevId: 191660378
* Adding Operation._control_outputsGravatar Alexandre Passos2018-04-04
| | | | PiperOrigin-RevId: 191659944
* Remove dependency on absl to fix tf-nightly-gpu windows GPU buildGravatar Smit Hinsu2018-04-04
| | | | PiperOrigin-RevId: 191659091
* Fix typo in `tf.reduce_mean` documentation.Gravatar Daniel Zheng2018-04-04
| | | | | | The range of valid values for the `axis` argument should be `[-rank(input), rank(input))`, just like other reduction ops. PiperOrigin-RevId: 191657767
* Iterate over a copy of dictionary keys when closing variable subscopes. ↵Gravatar Priya Gupta2018-04-04
| | | | | | Otherwise, we run into a "dictionary changed size during iteration" once in a while, as we are modifying the values in the dictionary during the iteration. PiperOrigin-RevId: 191655599
* Creates a LinearModel (inherits from keras.training.Model) that creates a linearGravatar Rohan Jain2018-04-04
| | | | | | | | model. Had to modify the __call__ method in the base layer class so that it could work with feature style inputs in which case we lazily convert the inputs to tensors instead of providing tensors as inputs upfront. PiperOrigin-RevId: 191655445
* Rename `distribute` to `train_distribute` parameter in `RunConfig` to ↵Gravatar Priya Gupta2018-04-04
| | | | | | clarify that its purpose is only for training. PiperOrigin-RevId: 191654161
* Add quantized uint8 L2Normalization Kernel.Gravatar A. Unique TensorFlower2018-04-04
| | | | PiperOrigin-RevId: 191652174
generated by cgit v1.2.3 (git 2.39.1) at 2024-08-06 23:04:29 +0000