aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Adding HighwayWrapper for rnn cell that creates a highway skip connection ↵Gravatar A. Unique TensorFlower2017-05-01
| | | | | | between the cell's input and output activations. Change: 154741723
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-05-01
| | | | Change: 154738165
* call InitMain in hvx ops support checkerGravatar A. Unique TensorFlower2017-05-01
| | | | Change: 154736553
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-05-01
| | | | Change: 154733347
* Move bucketize op to math_ops. This exposes the method as ↵Gravatar A. Unique TensorFlower2017-05-01
| | | | | | math_op._bucketize, but does not expose it in tf public interface. The method is still exposed as tf.contrib.layers.bucketize. Change: 154731971
* Enabling python configuration to use a remotely generated configuration. ↵Gravatar A. Unique TensorFlower2017-05-01
| | | | | | This will prepare TF for use of Bazel's new remote execution api. Change: 154726835
* Remove outdated README.Gravatar Martin Wicke2017-05-01
| | | | Change: 154726428
* Update "Using TensorFlow via Docker" README to mention memory limits.Gravatar Skye Wanderman-Milne2017-05-01
| | | | | Fixes #7861 Change: 154721550
* tfdbg: improve the way in which the no-debugger-linked error is raisedGravatar Shanqing Cai2017-05-01
| | | | Change: 154720434
* Make FunctionLibraryDefinition::AddFunctionDef() check for conflicting op nameGravatar Skye Wanderman-Milne2017-05-01
| | | | | This prevents a function from masking an existing op. Change: 154720287
* Removing unnecessary cuda:platform.bzlGravatar A. Unique TensorFlower2017-05-01
| | | | Change: 154712918
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-05-01
| | | | Change: 154711571
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-05-01
| | | | Change: 154711274
* Docs fix for sdca_ops.ccGravatar Mark Daoust2017-05-01
| | | | Change: 154710766
* Clarify docs on unused GraphKeys:Gravatar Mark Daoust2017-05-01
| | | | | | | | | WEIGHTS, BIASES, ACTIVTIONS Closes #7741 Change: 154708825
* Add ReferenceUtil::ApplyElementwise2D.Gravatar Justin Lebar2017-04-30
| | | | Change: 154687238
* Details about performance models and techniquesGravatar Toby Boyd2017-04-30
| | | | Change: 154679628
* [XLA] Make ReshapeMover account for broadcast operands, add VLOGging for debug.Gravatar A. Unique TensorFlower2017-04-29
| | | | Change: 154637127
* Add TFDBG support to GrpcSessionGravatar Shanqing Cai2017-04-29
| | | | | | | | * Along the way, unify the way the debugger works in DirectSession (non-distributed Sessions) and MasterSession (for distributed Sessions). * The SummarizDebugTensorWatches method is invoked in DirectSession::GetOrCreateExecutors() and MasterSession::HashBuildGraphOptions() method to generate keys for partition graphs and executors. * The DebugStateInterface::PublishDebugMetadata() method is used to send metadata about the debugged Session::Run() call to debug URLs. This happens in DirectSession::Run() and MasterSession::DoRunWithLocalExecution() respectively. * The DebugGraphDecoratorInterface::DecorateGraph() and DebugGraphDecoratorInterface::PublishGraph() methods are used to insert debug ops to the debugged graph and send the modified graph to debug URLs. This happens in DirectSession::GetOrCreateExecutors() and GraphMgr::InitItem(), respectively. Change: 154631802
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154614993
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154614872
* Fix broken links to images, make all image links absolute.Gravatar Martin Wicke2017-04-28
| | | | | | | Fixes #8064, fixes #7685. (after docs republish) Change: 154614227
* Refactor training helper functions to separate library.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154613254
* Distinguish between duplicate feed/fetch and unspecified feed/fetch errors.Gravatar Suharsh Sivakumar2017-04-28
| | | | Change: 154606429
* Added graph structure output to summarize_graphGravatar Pete Warden2017-04-28
| | | | Change: 154606362
* Actually, we may change the type of errors too.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154605805
* Internal Change.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154605009
* ReferenceUtil::ReduceWindow4DGeneric() now supports arbitrary edge padding.Gravatar Bjarke Hammersholt Roune2017-04-28
| | | | Change: 154603866
* When allocating GPU constants, check to see if the destinationGravatar Vijay Vasudevan2017-04-28
| | | | | | | | tensor is intialized early (because we ran out of memory) and report it as such. Fixes #7025. Change: 154603030
* Fixup tf.contrib.labeled_tensor.select to handle tuples properlyGravatar Stephan Hoyer2017-04-28
| | | | | | You can use a tuple to label a point along an axis, so tuples should not be converted into lists of points for indexing. Change: 154603014
* Add bool field inaccurate to Costs.Gravatar Yao Zhang2017-04-28
| | | | Change: 154602604
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154598657
* Remove or warn about nn_ops accidentally registered to intsGravatar Geoffrey Irving2017-04-28
| | | | | | | | | | | Several nn_ops were registered for ints accidentally. Some of these even have kernel registrations for ints which do very strange things. The ops without kernels registered have been restricted to floating point types, and warnings have been added for those with kernels. Other ops were registered with more float types than they had kernels for. Fixes #9317. Fixes #5539. Change: 154595629
* Document that tf.train.batch is nondeterministicGravatar Geoffrey Irving2017-04-28
| | | | | Fixes #9441. Change: 154593819
* Fix losses.get_regularization_losses documentation to avoid implying losses ↵Gravatar A. Unique TensorFlower2017-04-28
| | | | | | are Variables. Change: 154593246
* [XLA] Make HLO ordering module-scoped.Gravatar Mark Heffernan2017-04-28
| | | | | Add comparison of ordering of HLO instructions which are in different computations using the call graph. Previously, instructions in different computations were considered unordered. Ordering these instructions improves buffer liveness analysis and may enable better buffer sharing between values in different computations. Change: 154592912
* Added expalanation when to use embedding_lookup_sparse instead of ↵Gravatar Mustafa Ispir2017-04-28
| | | | | | | sparse_tensor_dense_matmul. Fixes #8131 Change: 154591681
* Documentation fixes for TF Layers tutorial (see #8301)Gravatar Neal Wu2017-04-28
| | | | Change: 154591046
* Improve docstring on strided_slice to refer users to easier function.Gravatar Andrew Selle2017-04-28
| | | | | Fixes #5761 Change: 154590417
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154590336
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154589265
* Update documentation to drop GPU support on osx for 1.2Gravatar Alexandre Passos2017-04-28
| | | | Change: 154589003
* Clarify candidate sampling ops documentationGravatar A. Unique TensorFlower2017-04-28
| | | | | | | Removes incorrect "per batch" language. Fixes #3824. Change: 154587896
* Documented naming insconsistence of variable_scope where reuse on and other ↵Gravatar Mustafa Ispir2017-04-28
| | | | | | ops doesn't share the scope. Change: 154586285
* Small comment fix.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154586226
* Support label_keys in DNNLinearCombinedClassifier and in LinearClassifier.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154585848
* Adds support for repeated images when parsing examples.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154584953
* Improves shared_name documentation for string_input_producerGravatar Alexandre Passos2017-04-28
| | | | Change: 154584844
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154584479
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-04-28
| | | | Change: 154583581