aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
...
* Shorten docstring line.Gravatar A. Unique TensorFlower2017-07-26
| | | | PiperOrigin-RevId: 163269709
* Fix comment ypoGravatar Asim Shankar2017-07-26
| | | | PiperOrigin-RevId: 163266376
* [XLA] Fix a bug in cloning outfeeds, carried the wrong shape.Gravatar Chris Leary2017-07-26
| | | | PiperOrigin-RevId: 163265592
* Rollback of GPU kernel implementation of transpose for tensors with one ↵Gravatar Yangzihao Wang2017-07-26
| | | | | | | | | | | small dimension. END_PUBLIC BEGIN_PUBLIC Automated g4 rollback of changelist 162525519 PiperOrigin-RevId: 163265445
* Set device for nodes added by LayoutOptimizer.Gravatar Yao Zhang2017-07-26
| | | | PiperOrigin-RevId: 163263257
* [XLA] Refactor insert-ReducePrecisionInsertion-pass code into a separate ↵Gravatar A. Unique TensorFlower2017-07-26
| | | | | | function to simplify optimization functions. PiperOrigin-RevId: 163250247
* Build fix for Android nightly buildGravatar A. Unique TensorFlower2017-07-26
| | | | PiperOrigin-RevId: 163249287
* [tpu:profiler] Make duration of tracing changeable.Gravatar Eric Liu2017-07-26
| | | | | | | Make the default duration 2 seconds. Also avoid dumping empty trace to log directory. PiperOrigin-RevId: 163248374
* Reflow comments to avoid being too long.Gravatar A. Unique TensorFlower2017-07-26
| | | | PiperOrigin-RevId: 163246389
* Fix arg lists for args and kwargs in docs.Gravatar Mark Daoust2017-07-26
| | | | PiperOrigin-RevId: 163246117
* Updated the memory optimization config to introduce an explicit default ↵Gravatar Benoit Steiner2017-07-26
| | | | | | value. This will make it possible change change the default behavior in the future by updating the meta optimizer code to interpret that default value differently (e.g we could assume default means heuristics). The default value remains OFF. PiperOrigin-RevId: 163239483
* Introduced a default setting for constant folding, currently set to OFF. WillGravatar Benoit Steiner2017-07-26
| | | | | | be turned to on later on. PiperOrigin-RevId: 163233994
* Setup a resource manager and enable the use of multiple threads during foldingGravatar Benoit Steiner2017-07-26
| | | | PiperOrigin-RevId: 163233960
* Add LocalTempFilename function to Env class which creates a local temp file nameGravatar Yuefeng Zhou2017-07-26
| | | | PiperOrigin-RevId: 163230810
* Create new tf.learn.estimator.head which is based on a tf.DistributionGravatar Joshua V. Dillon2017-07-26
| | | | | | instance. PiperOrigin-RevId: 163213141
* [XLA] Finish normalizing fusion computations into standard computationsGravatar A. Unique TensorFlower2017-07-26
| | | | PiperOrigin-RevId: 163210327
* Replace local tag with requires-gpu tag for cuda and sycl testsGravatar Shanqing Cai2017-07-26
| | | | PiperOrigin-RevId: 163205190
* New "Estimators" unit of Programmer's Guide.Gravatar A. Unique TensorFlower2017-07-26
| | | | PiperOrigin-RevId: 163189681
* DirectSession: Add accessor to DeviceMgr.Gravatar Asim Shankar2017-07-26
| | | | | | | This will enable some experimentation with keeping TF_Tensor objects in the C API backed by device memory (avoiding copies to host memory). PiperOrigin-RevId: 163173902
* Automated g4 rollback of changelist 163164566Gravatar Kay Zhu2017-07-25
| | | | PiperOrigin-RevId: 163170549
* Removing session reset since destroying the session object would delete its ↵Gravatar Yuefeng Zhou2017-07-25
| | | | | | variables as well. Resetting session might unintentionally close other sessions in the same process. PiperOrigin-RevId: 163164738
* [XLA] Use HloEvaluator for convolution in reference_util.Gravatar Kay Zhu2017-07-25
| | | | PiperOrigin-RevId: 163164566
* Internal change.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163160761
* Automated g4 rollback of changelist 163128673Gravatar Yuefeng Zhou2017-07-25
| | | | PiperOrigin-RevId: 163157277
* Plumbs rendezvous through function calls.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163157033
* Implemented MemoryUsed function for hash tables.Gravatar Yuefeng Zhou2017-07-25
| | | | PiperOrigin-RevId: 163149969
* `.get_shape()` -> `.shape` in input.py.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163147646
* Call AddDefaultAttrsToGraphDef() in grappler_item_builder.cc with op_registryGravatar A. Unique TensorFlower2017-07-25
| | | | | | | including function library so that it doesn't fail for the graph with inline function library. PiperOrigin-RevId: 163142677
* Use `to_float` and `to_int32` where appropriate.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163138030
* Add GDN activation function.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163137487
* Prune trivial ops (concatenation of a single tensor, AddN of a single tensor,Gravatar Benoit Steiner2017-07-25
| | | | | | ...) PiperOrigin-RevId: 163131793
* Add LocalTempFilename function to Env class which creates a local temp file nameGravatar Yuefeng Zhou2017-07-25
| | | | PiperOrigin-RevId: 163128673
* Make fft_length optional for inverse_stftGravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163127500
* Support customized residual function in the residual wrapper.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163121296
* [TF:XLA] Ignore control edges from Enter nodes to the graph sink during loop ↵Gravatar Peter Hawkins2017-07-25
| | | | | | functionalization. PiperOrigin-RevId: 163115904
* Convert configure to python.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163114551
* Improved the speed of constant foldingGravatar Benoit Steiner2017-07-25
| | | | PiperOrigin-RevId: 163113085
* Add gradient checking tests for nn.moments().Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163110994
* Implementation of UnsortedSegmentSum in tf2xla bridge.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163109769
* Remove duplicate import.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163108237
* Pass batch_size in params when use_tpu=False.Gravatar Jonathan Hseu2017-07-25
| | | | PiperOrigin-RevId: 163105673
* Adding missing deps to targets in llvm.BUILD. This was only working in ↵Gravatar A. Unique TensorFlower2017-07-25
| | | | | | non-sandboxed builds. PiperOrigin-RevId: 163103908
* [XLA] Add more unit tests for DynamicSlice and DynamicUpdateSlice.Gravatar Peter Hawkins2017-07-25
| | | | PiperOrigin-RevId: 163102445
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163102070
* [BatchNormGrad] Add end-to-end test for BatchNormGradGravatar A. Unique TensorFlower2017-07-25
| | | | | RELNOTES: n/a PiperOrigin-RevId: 163101568
* Minor tweaks to avoid unnecessary copiesGravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163101160
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-07-25
| | | | PiperOrigin-RevId: 163094455
* Add new CompareAndBitpackOp.Gravatar Eugene Brevdo2017-07-25
| | | | PiperOrigin-RevId: 163093146
* Speed up tf.contrib.signal spectral_ops_test.py by reducing the size of the ↵Gravatar RJ Ryan2017-07-25
| | | | | | gradient test. PiperOrigin-RevId: 163092423
* C API: Fix a bug with TF_OperationGetAttrTensor when TF_STRING tensors areGravatar Asim Shankar2017-07-25
| | | | | | | | | | | | | | involved. The TensorBuffer owned by a TF_Tensor object has a different memory layout than the TensorBuffer owned by the corresponding tensorflow::Tensor object. This change consolidates conversions between the runtime's tensorflow::Tensor and the C API's TF_Tensor objects into a pair helper functions. The added test: CApiAttributesTest.StringTensor fails without corresponding changes to c_api.cc PiperOrigin-RevId: 163091789