aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Add dummy function and data for inception inference on HVXGravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144763407
* In ./configure, only ask about jemalloc when running on Linux.Gravatar Jonathan Hseu2017-01-17
| | | | | | | | Also add functions to test whether we're running on Linux or macOS, and convert every check over to them. Fixes #6869 Change: 144761341
* Always linkstatic when building rpcbench_test.Gravatar Jonathan Hseu2017-01-17
| | | | Change: 144760356
* VariableOp shouldn't own a reference to state stored in the ResourceManager ↵Gravatar A. Unique TensorFlower2017-01-17
| | | | | | as that state can be cleared. Change: 144760207
* Add Makefile for hexagon controllerGravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144755322
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144754803
* Go: Support setting shape valued attributes.Gravatar Asim Shankar2017-01-17
| | | | | Fixes #6833 Change: 144752893
* Rename `weights` in `Dense` layer to `kernel`, and add base layer aliases ↵Gravatar Francois Chollet2017-01-17
| | | | | | for `(non_)trainable_weights`. Change: 144752883
* Add a new lightweight queue-like object - RecordInputGravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144752664
* Remove extra unnecessary parameter to functor::ResizeBilinearGravatar Brennan Saeta2017-01-17
| | | | | | | This extra parameter got introduced in a previous commit as part of some performance improvements. But because it is unused, it should be reverted back. Change: 144749662
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144749626
* Update C++ API comments to be more Doxygen friendly.Gravatar Skye Wanderman-Milne2017-01-17
| | | | | | | | | | | | This patch: - Updates // comments to ///. I manually reverted some comments that shouldn't be docs (e.g. TODOs), but may have missed some. - Indents code blocks so they get formatted as such in the docs. - Removes /* */ comments from example code since it messes up Doxygen. - Puts a space between {{ and }} since it messes up devsite. - Adds some // START_SKIP_DOXYGEN and // END_SKIP_DOXYGEN comments for functions that aren't part of the public API (incomplete) This will likely require further small fixups, but this gets something to be generated. Change: 144749351
* Fix error message in direct_session.Gravatar Suharsh Sivakumar2017-01-17
| | | | Change: 144749245
* Moved FeedFnHook into basic_session_run_hooks.Gravatar Illia Polosukhin2017-01-17
| | | | | | | Make LoggingTensorHook to print tensors in the order they were given [if list]. Added formatter option, to support custom string formatting. Added numpy printing options configuration to tweak precision and summarization. Change: 144748847
* Simplifications and cleanups in the implementation of ParallelConcatGravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144747557
* Add experimental implementation for hexagon controllerGravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144746814
* Add several operations to the StreamExecutor APIGravatar A. Unique TensorFlower2017-01-17
| | | | | No implementations are yet provided for these operations. Change: 144743665
* Plumb port::Status through the internal synchronous memcopy routines.Gravatar A. Unique TensorFlower2017-01-17
| | | | | | | Now, at least for the public APIs that return port::Status, they can grab the port::Status that the implementation would like to return and use its additional information in reporting to the user. Change: 144741667
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144739269
* Fix debian build.Gravatar Jan Prach2017-01-17
| | | | Change: 144738033
* [XLA] Privatize `HloInstruction::CreateNary`, NFCGravatar A. Unique TensorFlower2017-01-17
| | | | | This method is only used in other `HloInstruction` creator methods. Change: 144737654
* Make dtype consistent with docs for TP/TN/FP/FN ops.Gravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144737436
* Turn on GPU growth option by default for InteractiveSession if no configGravatar Vijay Vasudevan2017-01-17
| | | | | provided. This is probably a better default to have. Change: 144737122
* Sync header file to match the implementations.Gravatar Brennan Saeta2017-01-17
| | | | Change: 144734632
* Setting the default axis to 0 for argmax and argminGravatar Manjunath Kudlur2017-01-17
| | | | Change: 144733955
* Initial design of new seq2seq Decoder class library.Gravatar Eugene Brevdo2017-01-17
| | | | | | | | Still hidden from tf.contrib.seq2seq namespace for rapid iteration. Not for general use until the API has stabilized a bit. Barebones documentation for now. Change: 144733714
* Restrict weights rank to be the same as the broadcast target, to avoid ↵Gravatar A. Unique TensorFlower2017-01-17
| | | | | | ambiguity on broadcast rules. Change: 144733601
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144733155
* Merge changes from github.Gravatar Patrick Nguyen2017-01-17
| | | | Change: 144729490
* Name change in LinearOperator: batch_shape_dynamic --> batch_shape_tensor.Gravatar Ian Langmore2017-01-17
| | | | | Similarly for other "dynamic" Ops. Change: 144728885
* Adding ValueError checking for summaries argument of optimize_loss (similar ↵Gravatar A. Unique TensorFlower2017-01-17
| | | | | | to optimizer name checking) Change: 144727719
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144726940
* Improve the performance of resize_bilinear.Gravatar Brennan Saeta2017-01-17
| | | | | | | | | | | | | | | | | | The previous implementation of resize_bilinear operation is a straight forward, naive implementation. Unfortunately, it leaves a significant amount of room for improvement. This change improves the performance of the operation by avoiding unnecessary recomputation (for where the operation was CPU bound), and avoids redundant memory references (for where the operation was memory bound). In my benchmark (loosely based on inception image pipeline), this implementation change improves speed between 1.5-2x. Additionally, to ensure correctness, I've preserved the old implementation of resize_bilinear, and added a number of tests that initalize a random image, and ensure that the outputs are identical. Change: 144726607
* Added summary-save-sec option to the MonitoredTrainingSession, which is ↵Gravatar Mustafa Ispir2017-01-17
| | | | | | commonly used in example models. Change: 144721588
* Add a micro-benchmark for the resize_bilinear_op to the image_ops_test.Gravatar Brennan Saeta2017-01-17
| | | | | | | | | | | Based on instrumentation of the imagenet/inception_train typical input & output image sizes, we have devised a micro benchmark to verify the performance of the resize_bilinear op. To compile and run: - bazel build -c opt --copt=-mavx //tensorflow/python/image_ops_test - ./bazel-bin/tensorflow/python/image_ops_test --benchmarks=ResizeBilinear Change: 144718335
* [TF:XLA] Mark _XlaLaunch as a stateful op since XLA's random-number ↵Gravatar Peter Hawkins2017-01-17
| | | | | | | | | generators are stateful. Prevents Tensorflow from constant-folding _XlaLaunch ops containing random-number ops. Fixes Github issue: #6854 Enable RandomStandardNormal for the XLA CPU backend. Change: 144716342
* tf.Learn export: leave serving batch size unspecified by defaultGravatar David Soergel2017-01-17
| | | | Change: 144713722
* Improved error report when importing a sub-meta-graphdef with missing inputs.Gravatar A. Unique TensorFlower2017-01-17
| | | | Change: 144689500
* Use integer division to compute number of updates.Gravatar Patrick Nguyen2017-01-16
| | | | Change: 144675800
* Automated rollback of change 144470928Gravatar Patrick Nguyen2017-01-16
| | | | Change: 144673014
* Remove tf.cloud from public API until it's ready.Gravatar Patrick Nguyen2017-01-16
| | | | Change: 144669682
* Fixed string formatting and valid task listing.Gravatar A. Unique TensorFlower2017-01-16
| | | | Change: 144656658
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2017-01-16
| | | | Change: 144648190
* tfdbg: Add GrpcDebugWrapperSessionGravatar Shanqing Cai2017-01-16
| | | | | | | The newly-added GrpcDebugWrapperSesion and the existing DumpingDebugWrapperSession now share a same superclass, namely the newly-created NonInteractiveDebugWrapperSession in python/debug/wrappers/framework.py. GrpcDebugWrapperSession is not yet exposed in the tensorflow.python.debug namespace, for rapid iteration reasons. It'll be made public once its API has stabilized together with grpc_debug_server. Change: 144647632
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2017-01-16
| | | | Change: 144623381
* Updated bucket_ops to support different batch sizes per bucket.Gravatar A. Unique TensorFlower2017-01-16
| | | | | (can be used to have larger batch sizes for smaller input sequences.) Change: 144621786
* Create --config opt for compiling fully optimized binaries.Gravatar Gunhan Gulsoy2017-01-16
| | | | Change: 144609556
* Adding nav entries for Layers tutorial, and making a few small formatting ↵Gravatar Sanders Kleinfeld2017-01-15
| | | | | | fixes to it. Change: 144588235
* Fix a breakage in python3 testGravatar Shanqing Cai2017-01-15
| | | | Change: 144584357
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2017-01-15
| | | | Change: 144575750
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-07 22:33:42 +0000