aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
...
* Internal changes.Gravatar A. Unique TensorFlower2017-04-11
| | | | Change: 152876919
* Add functions to tfdbg for parsing time and memory intervals.Gravatar A. Unique TensorFlower2017-04-11
| | | | Change: 152876511
* N/AGravatar A. Unique TensorFlower2017-04-11
| | | | Change: 152874742
* tfdbg: fix bug related to compiled Python filesGravatar Shanqing Cai2017-04-11
| | | | | Previously, the modules source_utils.py and analyzer_cli.py did not take into account that files in the stack could be compiled (i.e., pyc and pyo files). Change: 152873179
* Relabeling some tests.Gravatar Vinu Rajashekhar2017-04-11
| | | | | | - Both of these are at the edge of their allowed limits in terms of test timeout. Change: 152873079
* Add AudioSpectrogram op to TensorFlow for audio feature generationGravatar Pete Warden2017-04-11
| | | | Change: 152872386
* fix formatting errors and broken links inside "Getting Started With TensorFlow"Gravatar A. Unique TensorFlower2017-04-11
| | | | Change: 152867721
* Backport fixes and improvements from external Keras.Gravatar Francois Chollet2017-04-11
| | | | Change: 152867002
* Docs: Tutorial formatting and links fixGravatar Asim Shankar2017-04-11
| | | | | | | Fixes #9092 (When accompanied by changes to the www.tensorflow.org/code redirects to point to the appropriate release branch on github) Change: 152866886
* Update the docstring of tuner.Gravatar Jianwei Xie2017-04-11
| | | | Change: 152866121
* Add tf.unsorted_segment_sum GPU kernel for complex64/complex128.Gravatar RJ Ryan2017-04-11
| | | | Change: 152864723
* Deprecate PlaceholderV2 since we found a way to make the same changeGravatar Vijay Vasudevan2017-04-11
| | | | | to Placeholder in a backwards compatible way. Change: 152843957
* Support two data types as inputs in RemoteFusedGraphExecuteOpGravatar A. Unique TensorFlower2017-04-11
| | | | Change: 152843285
* Allow passing custom hooks to 'evaluation_loop'.Gravatar Vijay Vasudevan2017-04-11
| | | | Change: 152838415
* SimpleGraphExecutionState::InitBaseGraph(): don't pass ↵Gravatar Skye Wanderman-Milne2017-04-11
| | | | | | | FunctionLibraryDefinition to Graph ctor ConvertGraphDefToGraph() adds functions to the graph now, so there's no need to explicitly pass them into the graph constructor. Change: 152837855
* Gracefully exit on writing config file.Gravatar A. Unique TensorFlower2017-04-11
| | | | Change: 152837817
* Update release notes with mention of `tf.Session.make_callable()`.Gravatar Derek Murray2017-04-11
| | | | Change: 152837661
* Adding license headers to cmake files.Gravatar A. Unique TensorFlower2017-04-11
| | | | Change: 152835531
* Fix QuantizedToFloat to avoid bias.Gravatar A. Unique TensorFlower2017-04-11
| | | | Change: 152835355
* Add a tf.neg GPU kernel for complex64/complex128.Gravatar RJ Ryan2017-04-11
| | | | | Also, enable GPU tests for ops that are already supported on GPU. Change: 152831861
* [XLA] Fix cloning of tuples in HloInstruction::CloneWithNewOperands.Gravatar A. Unique TensorFlower2017-04-11
| | | | | Tuple instructions would previously not be cloned with new operands. Change: 152830440
* Fixes tf.cond with nested return values inside while loops.Gravatar A. Unique TensorFlower2017-04-11
| | | | Change: 152830112
* Enable TensorArray GPU kernels for complex64/complex128.Gravatar RJ Ryan2017-04-11
| | | | Change: 152824547
* Using GMock matchers in XLA tests.Gravatar A. Unique TensorFlower2017-04-11
| | | | Change: 152823724
* Remove explicit FunctionLibraryDefinition from tfcompile_main.ccGravatar Skye Wanderman-Milne2017-04-11
| | | | | Instead of passing around a manually-constructed FunctionLibraryDefinition, use the constructed graph's FunctionLibraryDefinition. Change: 152821479
* [XLA:CPU] Disable AVX512Gravatar David Majnemer2017-04-11
| | | | | | The AVX512 support in LLVM hasn't solidified enough. Ignore support for it while it stabilizes. Change: 152819598
* [XLA:CPU] VLOG(1) the target CPU and its feature in SimpleOrcJIT's constructor.Gravatar Peter Hawkins2017-04-11
| | | | Change: 152802749
* Allow callers to control the names of the slot variables.Gravatar A. Unique TensorFlower2017-04-11
| | | | | | | | | Other optimizers which have only a single slot variable allow control through the 'name' constructor parameter, but the FtrlOptimizer has two variables. Because they both are created with the same 'name' parameter, one of them has name as a suffix, and the other has name + "_1" as a suffix. This change allows them to be specified in a more controllable way. Change: 152802478
* Add a utility to check ops in a graph are supported on hvxGravatar A. Unique TensorFlower2017-04-10
| | | | Change: 152768777
* Add Split and SplitV GPU support for complex64/complex128.Gravatar RJ Ryan2017-04-10
| | | | | Update split_op_test to test complex numbers in addition to floats and always test with a GPU if one is available. Change: 152757648
* Adds `tf.Session.make_callable()`.Gravatar Derek Murray2017-04-10
| | | | | | | | | | | | | | | | | | | This method returns a Python callable that has the same semantics as `tf.Session.run()`, but can cache some of the work that must be done to map Tensor-like objects to the arguments of the underlying C API function. The initial implementation is optimized for single-`Tensor` and single-`Operation` fetches, and delegates to `tf.Session.run()` for handling feeds. Since most queue runners use a single-`Operation` `run()` call, switch the `tf.train.QueueRunner` implementation to use `make_callable()` Using this new interface can improve the latency of small steps (measurements from my workstation): * The median time to fetch a 4-byte tensor decreases from 99us to 52us (-47us). * The median time to run a trivial op decreases from 80us to 31us (-49us). Change: 152757301
* Does a deep copy of the tensors output from GraphRunner::Run(...)Gravatar Vinu Rajashekhar2017-04-10
| | | | | | | | | | | The current GraphRunner::Run(...) outputs tensors produced from running the Executor on the graph, but these tensors are actually owned by the allocator from the device created for the Run(...), which could be deleted along with the device. The deep copy allows the ownership to be transferred to the global static cpu_allocator(). Before, the allocator was always a global cpu_allocator(), but with a recent change there is an option to tie allocations to a memory limited allocator per-session. Change: 152756520
* Add support for solvers based on the Nvidia cuSolver library.Gravatar A. Unique TensorFlower2017-04-10
| | | | | Implement a GPU version of tf.cholesky as a first example. Change: 152756373
* Add tf.concat GPU kernels for complex64/complex128.Gravatar RJ Ryan2017-04-10
| | | | | Also update concat_op_test to always use_gpu=True where previously it was running each test twice. Change: 152755973
* Add option to return loss as batch sum, or divided by number of non-zero ↵Gravatar A. Unique TensorFlower2017-04-10
| | | | | | weights. Change: 152754313
* [tf contrib seq2seq] Change 'attention_history' to 'alignment_history'Gravatar Eugene Brevdo2017-04-10
| | | | | (this is the history most folks want; and attention is returned as an output). Change: 152752862
* Change Placeholder to support partial shapes and enforce scalar shapes.Gravatar Vijay Vasudevan2017-04-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Adds tests; testScalar failed before with the original placeholder because it treated [] as "?" instead of scalar. Now you can actually specify [] and it means 'scalar'. Added a backwards compatibility test using a graph_def generated from a previous tf version. RELNOTES: tf.placeholder can represent scalar shapes and partially known shapes accurately. Note, this change can break already buggy programs because it makes placeholder shape handling more consistent across graph serializations. Note: There are some buggy cases where this change can break a buggy pipeline: namely those that serialize a graph using an unknown shape (e.g., [None, 10] in a tf.placeholder, but then reload the graph using import_graph_def and feed it a different shape. Prior to this change, serializing the graph_def loses the [None, 10] shape requirement, so you can feed anything you want. This change makes it so that you serialize the graph with [None, 10], and so when you reload it, it would fail if you fed it a different shape. In these cases, the fix is to correct the original placeholder shape to match what you feed it, which is not a bug in TF but in the user's program. Note 2: A python user that did tf.placeholder(shape=[]) would get scalar checking in the same process due to python shape inference code. However, a C++ user that did Placeholder(shape=[]) would not have gotten scalar shape checking; a C++ program that passed Placeholder(shape=[]) that expects to interpret this as "UnknownShape" would break -- however, that user could have already used an {unknown_shape: true} proto, and should not have expected the legacy behavior. Backwards compatibility: Old graphs that have shape = {} in the proto will also have a graph_def_version <= 21, so the default value of shape prior to this change will be interpreted by new binaries as "UnknownShape" just as before. Forwards compatibility: new graphs will produce, by default, shape={ unknown rank: true}; old binaries will use PartialTensorShape's parsing code to parse that proto into an object whose shape.dims() <= 0, and so these binaries will continue to interpret the default shape as "unknown shape" without crashing and without producing new errors. Fixes #9103 Change: 152751019
* Fix some size_t casting issues.Gravatar Suharsh Sivakumar2017-04-10
| | | | Change: 152749005
* Fix a segfault in AlgebraicSimplifierVisitor.Gravatar Jonathan Hseu2017-04-10
| | | | Change: 152743270
* Add a missing "const" to an operator- method.Gravatar A. Unique TensorFlower2017-04-10
| | | | | Found by running unit-tests with GLIBCXX_DEBUG. Change: 152742124
* Add run_mode field to TestResults proto for differentiating between test types.Gravatar Andrew Harp2017-04-10
| | | | Change: 152740961
* Allows summary writer to take an optional suffix= argument, whichGravatar A. Unique TensorFlower2017-04-10
| | | | | is appended to every summary file name. Change: 152736272
* tfdbg: disable problematic watched nodes under file:// debug URL in ↵Gravatar Shanqing Cai2017-04-10
| | | | | | testDebugTrainingDynamicRNNWorks Change: 152735279
* tfdbg core: allow gRPC debug server to remotely disable/enable debug opsGravatar Shanqing Cai2017-04-10
| | | | | | | | | | | | | | | Synopsis of changes: * The EventReply protobuf is expanded: a new field called "debug_op_state_change" is added to allow the debug server to remotely enable and disable debug ops. * At the end of every debug gRPC stream, the server sends all the queued EventReply protos to the client. The client (i.e., the debugged TF runtime) receives them and toggle the enabled status of the debug ops accordingly. * Added gated_grpc attribute to existing debug ops. This new boolean attribute is set to False by default, ensuring backward compatibility in behavior. If set to True, the debug ops will send the output tensors through grpc:// streams if and only if they are currently enabled. Otherwise we say that the debug op is "gated off" at the grpc:// URL. * If a debug op is gated off at all URLs it possesses, it will perform no expensive computation and instead just emit an empty (size {0}) output tensor. Other detailed change descriptions: * All debug ops now share the same base class "BaseDebugOp" to reduce the amount of boilerplate, which has grown in size due to the new gRPC gating logic. Change: 152733779
* xla: Add filename to failure message in LoadIRModuleGravatar Asim Shankar2017-04-10
| | | | | Will help with issues like #8947 Change: 152733558
* Fix mac protobuf links.Gravatar Jonathan Hseu2017-04-10
| | | | | Fixes #9089 Change: 152732901
* Choosing a saner default for min_eval_frequency in the constructor for ↵Gravatar A. Unique TensorFlower2017-04-10
| | | | | | Experiment for the GCS file system, because the default of 1 causes performance problems. Change: 152731761
* [TF:XLA] Don't clear the device name on nodes in graph_to_functiondef.cc.Gravatar Peter Hawkins2017-04-10
| | | | Change: 152730792
* Support partitioned bias variable in sampled_softmax_loss.Gravatar A. Unique TensorFlower2017-04-10
| | | | Change: 152730499
* Updating update_version to update the suffix in core.hGravatar A. Unique TensorFlower2017-04-10
| | | | Change: 152729057
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-09 17:36:28 +0000