aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Add ExportRunMetadata in queue runner and ExportCostGraph in coordinator.Gravatar Yuefeng Zhou2017-03-17
| | | | | Make the queue runner own the metadata and mutex. Change: 150475730
* [Tensorflow] Add check fail when user passes a tensor with nullptr to lookup.Gravatar A. Unique TensorFlower2017-03-17
| | | | Change: 150474503
* Android: Added download models into build.gradle for android exampleGravatar A. Unique TensorFlower2017-03-17
| | | | Change: 150471440
* [XLA] Add mapping from HloInstruction to CallGraphNode.Gravatar Mark Heffernan2017-03-17
| | | | | | | | | | Make mapping from calling instruction to CallSite one-to-one by extending CallSite to handle more than one called computation. This enables instructions like kWhile which call two computations to be represented as a single CallSite. Also add a mapping from instruction to CallSite in CallGraphNode to enable fast call site lookup. Also, include a few other opportunistic improvements: * Change CallGraph::Build factor to return a std::unique_ptr. This enables, for example, more convenient use of CallGraph as a data member to a class. * Change a few uses of unordered_set/map to FlatSet/Map. Change: 150469958
* Fix link in `constant_op.py`.Gravatar Mark Daoust2017-03-17
| | | | Change: 150466902
* Add an option to omit contrib tests in ci build.Gravatar Gunhan Gulsoy2017-03-17
| | | | Change: 150463151
* Adding the pip smoke test as the 8th test in sanity.Gravatar A. Unique TensorFlower2017-03-17
| | | | Change: 150462131
* Improve fused instruction dumpGravatar A. Unique TensorFlower2017-03-17
| | | | Change: 150460833
* Java: Constructor for creating Sessions with a ConfigProtoGravatar Asim Shankar2017-03-17
| | | | | | And also convenience functions for feeding and fetching using handles to outputs/operations instead of their names. Change: 150460697
* Remove the unnecessary RTLD_GLOBAL dlopen flag in the remaining tests.Gravatar Jonathan Hseu2017-03-17
| | | | Change: 150460215
* Added data type info to conv autotune parameters.Gravatar Yangzihao Wang2017-03-17
| | | | Change: 150459431
* Improves performance of tf.matmul(a, b, ...) for dense tensors on NVIDIA ↵Gravatar A. Unique TensorFlower2017-03-17
| | | | | | | | | | | | | | | | | | | | | | | | | | | GPUs in the following cases: a) If the inner-most dimension of b is 1, i.e. the operation is (possibly a batch of) matrix*vector multiplication(s). This is accomplished by calling Cublas GEMV rather than GEMM. This speeds up large matrix-vector products by about 4x. b) If one or more dimensions are unknown at graph construction time but the operation is in fact either a single matrix*matrix or matrix*vector multiplication. The following benchmark numbers illustrating the improvements for matrix * vector products were measured on a NVIDIA Titan X (Maxwell) card. Benchmark Base (ns) New (ns) Improvement ---------------------------------------------------------------------------- BM_Matmul_50_50_1_false_false_DT_FLOAT_gpu 18102 17056 +5.8% BM_Matmul_50_50_1_true_false_DT_FLOAT_gpu 18108 16374 +9.6% BM_Matmul_50_50_1_false_true_DT_FLOAT_gpu 18153 17173 +5.4% BM_Matmul_50_50_1_true_true_DT_FLOAT_gpu 18150 15950 +12.1% BM_Matmul_500_500_1_false_false_DT_FLOAT_gpu 64605 16874 +73.9% BM_Matmul_500_500_1_true_false_DT_FLOAT_gpu 62810 17298 +72.5% BM_Matmul_500_500_1_false_true_DT_FLOAT_gpu 60447 17014 +71.9% BM_Matmul_500_500_1_true_true_DT_FLOAT_gpu 58443 16934 +71.0% BM_Matmul_2000_2000_1_false_false_DT_FLOAT_gpu 343298 81898 +76.1% BM_Matmul_2000_2000_1_true_false_DT_FLOAT_gpu 294738 63723 +78.4% BM_Matmul_2000_2000_1_false_true_DT_FLOAT_gpu 300671 83650 +72.2% BM_Matmul_2000_2000_1_true_true_DT_FLOAT_gpu 284540 63742 +77.6% Change: 150456725
* Removes an unnecessary check that blocks using multihead with custom heads.Gravatar Zakaria Haque2017-03-17
| | | | Change: 150452316
* Set producer version in the graph used by shape refiner to run constantGravatar A. Unique TensorFlower2017-03-17
| | | | | folding. Change: 150450219
* Fix mis-spelling.Gravatar A. Unique TensorFlower2017-03-17
| | | | Change: 150450082
* [XLA] Add a test for the remainder of two scalar U32s.Gravatar A. Unique TensorFlower2017-03-17
| | | | Change: 150449788
* Copied global step tests from contrib to core.Gravatar Mustafa Ispir2017-03-17
| | | | Change: 150447439
* Tested features, labels, and mode in Estimator.exportGravatar Mustafa Ispir2017-03-17
| | | | Change: 150443246
* Use e.errno instead of trying to except FileExistsError.Gravatar A. Unique TensorFlower2017-03-17
| | | | | FileExistsError doesn't exist in Python 2.7. Change: 150436736
* - Update XLA for removal of TargetOptions::LessPreciseFPMADOption (LLVM r298023)Gravatar A. Unique TensorFlower2017-03-17
| | | | | - Removes getArgumentList() in favor of args() etc in XLA (LLVM r298010) Change: 150427660
* Updates to RNNCells to allow easy storage of attention TensorArray in the state.Gravatar Eugene Brevdo2017-03-17
| | | | | | | The main change is that RNNCells that wrap other RNNCells now override self.zero_state to call the wrapped cell's zero_state and then (maybe) perform some post-processing... instead of relying on the state_size property to provide all information about the state. Also made zero_state calls create ops inside their own name scope. Change: 150413265
* Initial cut of documentation for tf.contrib.seq2seqGravatar Eugene Brevdo2017-03-16
| | | | Change: 150400474
* Added an option to disable the collection of detailed statistics in grapplerGravatar Benoit Steiner2017-03-16
| | | | Change: 150397471
* tfdbg: Created a GRPC-based hook that streams debugger-related events.Gravatar A. Unique TensorFlower2017-03-16
| | | | Change: 150396376
* A simple script to test TF contrib.Gravatar Gunhan Gulsoy2017-03-16
| | | | Change: 150389857
* Sleep forever to trigger the timeout consistentlyGravatar Benoit Steiner2017-03-16
| | | | Change: 150388263
* Fix separable convolution bias checkGravatar A. Unique TensorFlower2017-03-16
| | | | Change: 150385615
* [Tensorflow] Expose API to lookup TensorSlice.Gravatar A. Unique TensorFlower2017-03-16
| | | | Change: 150384503
* Replace OpRegistryInterface* with FunctionLibraryDefinition in Graph.Gravatar Skye Wanderman-Milne2017-03-16
| | | | | This is a first step towards supporting functions in C++ graph construction, e.g. being able to import GraphDefs with functions. Change: 150382046
* Update opensource vulcanized HTML file for Tensorboard.Gravatar A. Unique TensorFlower2017-03-16
| | | | | This update contains refinements to the charts in the scalars dashboard. Change: 150380169
* Let the user view health pills at any step.Gravatar A. Unique TensorFlower2017-03-16
| | | | | | | This involves adding a toggle to the health pills info box in the graph visualizer. When that toggle is enabled, Tensorboard makes a request for health pills at step X when the user moves the slider. This feature can be very slow because it requires reading from disk. Viewing health pills at say step 100,000 could take minutes to an hour. We must design ways to make this faster (for instance, have the debugger write events at a much greater frequency only after it encounters a bad value). Change: 150379929
* Upgrade bazel to 0.4.5.Gravatar A. Unique TensorFlower2017-03-16
| | | | | | - Feature in 0.4.5 is required for cuda_configure to stay fix #7575 Change: 150376995
* Move the note for gcc version 5 to just below the bazel build command where ↵Gravatar Neal Wu2017-03-16
| | | | | | it actually applies Change: 150374176
* Allow soft placement on all the types of clusters.Gravatar Benoit Steiner2017-03-16
| | | | Change: 150373647
* Remove old doc generator.Gravatar Martin Wicke2017-03-16
| | | | Change: 150372607
* Smarter retry logic for non-idempotent file operations such as RenameFile, ↵Gravatar Alexey Surkov2017-03-16
| | | | | | DeleteFile or DeleteDir. Change: 150369708
* Change build visibility of tensorflow/compiler/tf2xla:xla_compilerGravatar A. Unique TensorFlower2017-03-16
| | | | Change: 150364996
* Autogenerated Change: Change TensorBoard TAG to 47Gravatar Dandelion Mané2017-03-16
| | | | Change: 150364914
* Automated rollback of change 150344647Gravatar Benoit Steiner2017-03-16
| | | | Change: 150363523
* Improve TensorBoard scalar dashboard domain calculation.Gravatar Dandelion Mané2017-03-16
| | | | | | | - Add an "ignore y-outliers" option (default true). When true, the domain is calculated based on the middle 8 deciles of the y range, i.e. the lowest 10% of data and highest 10% of data is ignored for domain calculation purposes. Also, this is done with the smoothed data rather than the raw data. This means that brief spikes or an initially high loss value will not distort the chart. This can be disabled to view the full data domain. - If the y values are all in the range [0, 1], then the domain is automatically set to be approximately [0, 1]. That makes it easy to compare proportional values. - Added bold gridlines at the origin (0,0) to make the charts easier to understand at a glance. Change: 150362432
* tfdbg: make debug_gateway_test a CPU-only test for nowGravatar Shanqing Cai2017-03-16
| | | | Change: 150361253
* Switch debug_grpc_testlib to bind on localhostGravatar A. Unique TensorFlower2017-03-16
| | | | | | | | Since there are no multi-machine tests, this is likely preferable from a security standpoint to binding to all interfaces, and sidesteps the question of cross-platform compatibility (some platforms may have IPV6_V6ONLY enabled by default). Change: 150358613
* Fix estimator tests when running against the installed pip package.Gravatar Jonathan Hseu2017-03-16
| | | | | This change allows the importing from tensorflow.python.estimator.* for usage in tests. Change: 150356830
* Add Dev Summit video.Gravatar A. Unique TensorFlower2017-03-16
| | | | Change: 150356236
* Forwards parallel_iterations arg in stack_bidirectional_dynamic_rnn().Gravatar A. Unique TensorFlower2017-03-16
| | | | Change: 150351473
* Embedded video in docs.Gravatar A. Unique TensorFlower2017-03-16
| | | | Change: 150348366
* Fixed a race condition in the single_cluster codeGravatar Benoit Steiner2017-03-16
| | | | Change: 150345708
* tfdbg: test examples and binaries against pip installGravatar Shanqing Cai2017-03-16
| | | | | | | I realized that CL/149810103 forgot to include the tfdbg binary and examples in the pip package. This CL fixes that and adds pip test for the correct inclusion of these files. Change: 150345489
* Disable flaky grappler:utils_testGravatar Gunhan Gulsoy2017-03-16
| | | | Change: 150344647
* Optimize LiteralUtil::Replicate and add a unit testGravatar A. Unique TensorFlower2017-03-16
| | | | Change: 150343323