aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
...
* Validate plugin names and routesGravatar A. Unique TensorFlower2017-05-24
| | | | | | | | | | | | | | A plugin name shouldn't contain slashes or be empty, as these would lead to confusing routes: a plugin with name "eggs/ham" and route "/data" would handle the route "/data/plugin/eggs/ham/data", but this route could also be handled by the plugin "eggs" along route "/ham/data". Furthermore, it could also be handled by the plugin with name "" along route "eggs/ham/data", or the plugin with name "eggs/h" along route "am/data". None of these is desirable, so I'm forbidding - plugins with characters not in [A-Za-z0-9_.-], and - routes without a leading slash. PiperOrigin-RevId: 157007010
* use 'global_gradient_norm' in summaries to enable only overall statsGravatar A. Unique TensorFlower2017-05-24
| | | | PiperOrigin-RevId: 157006018
* Fix bug when creating metrics variables inside a Tensorflow loop.Gravatar Peter Hawkins2017-05-24
| | | | PiperOrigin-RevId: 157005668
* Support tensors as dropout rates again, by removing the min(max(..)..) clippingGravatar A. Unique TensorFlower2017-05-24
| | | | | | that throws TypeError for tensors. PiperOrigin-RevId: 157000504
* Add metadata to generated doc pages.Gravatar Mark Daoust2017-05-24
| | | | PiperOrigin-RevId: 156990440
* Clean up tf_audio_dashboardGravatar Justine Tunney2017-05-23
| | | | | | | This change demonstrates my proposed "best practice" for having a documentation page, demo page, and a test package. PiperOrigin-RevId: 156945855
* tfdbg: increase the size of examples_test to mediumGravatar Shanqing Cai2017-05-23
| | | | | | | | To avoid test timeout in OSS builds such as: http://ci.tensorflow.org/view/Tensorflow%20Jenkins%20Monitored%20builds/job/nightly-matrix-cpu/TF_BUILD_IS_OPT=NO_OPT,TF_BUILD_IS_PIP=NO_PIP,TF_BUILD_PYTHON_VERSION=PYTHON3,label=mac-slave/494/console PiperOrigin-RevId: 156945796
* Disable test that is flakily failing on Mac OS X.Gravatar Derek Murray2017-05-23
| | | | PiperOrigin-RevId: 156944397
* Fix the issue in which pan to node would direct the user to some crazy ↵Gravatar A. Unique TensorFlower2017-05-23
| | | | | | location. The problem was that the new translateBy method introduced by d3 v4 applied scaling before translation. This change manually counters that with a division. PiperOrigin-RevId: 156943470
* Automated g4 rollback of changelist 156917266Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156939644
* Add graph transform rewriter to place arguments to fuse remote graphGravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156939364
* Register GPU kernels for bitcast.Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156939333
* Fix build dependency.Gravatar Yao Zhang2017-05-23
| | | | PiperOrigin-RevId: 156938197
* Fix the gpu time accounting.Gravatar A. Unique TensorFlower2017-05-23
| | | | | | | 1. Handle the case when node_name in RunMetadata is suffixed with :op_type. 2. Properly distinguish cpu/gpu time PiperOrigin-RevId: 156937421
* Remove references to Plottable having multiple versions.Gravatar Dandelion Man?2017-05-23
| | | | PiperOrigin-RevId: 156936949
* Output ES6 modules rather than NodeGravatar Justine Tunney2017-05-23
| | | | | | This will be useful if we want to run TensorBoard through the Closure Compiler. PiperOrigin-RevId: 156936819
* [XLA] In ReshapeMover: only sink Reshape/Transpose below elementwise HLO, if theGravatar Kay Zhu2017-05-23
| | | | | | | number of non-trivial reshape/transpose operands is non-zero. (Otherwise there's no benefit to sink Reshape/Transpose). PiperOrigin-RevId: 156936778
* Use _DefinedFunction.add_to_graph() call inside importer.Gravatar A. Unique TensorFlower2017-05-23
| | | | | | | | Before, Graph._add_function() call was used to import function library from GraphDef. This private function lacks hash value comparison and importing the inner function library used by the function gets improted. PiperOrigin-RevId: 156933480
* Reduce threshold of integration test to avoid flakes.Gravatar Francois Chollet2017-05-23
| | | | PiperOrigin-RevId: 156933321
* Created a regression test for the virtual placerGravatar Benoit Steiner2017-05-23
| | | | PiperOrigin-RevId: 156927675
* Tests _dnn_model_fn using mock head, modifies other tests to use consistent ↵Gravatar A. Unique TensorFlower2017-05-23
| | | | | | numbers, and drops some unnecessary tests. PiperOrigin-RevId: 156925729
* Fix regression in embeddings projectorGravatar Justine Tunney2017-05-23
| | | | | | This makes the ordering more reverse topological. PiperOrigin-RevId: 156924652
* Improve error message in cross_column re:hashing.Gravatar Mustafa Ispir2017-05-23
| | | | PiperOrigin-RevId: 156923445
* Added support for testing internal padding in R4.Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156919832
* Allow Estimator subclasses to override how the global step is created.Gravatar Jonathan Hseu2017-05-23
| | | | PiperOrigin-RevId: 156917877
* Fix `python/ops/losses/util.py` docstrings.Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156917717
* Fix oversampling in the GPU version of multinomial due to an error in generatingGravatar A. Unique TensorFlower2017-05-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | gumbel noise. -log(-log(U)) gives infinity if U draws a hard 0. Adds a tiny offset to U (2e-30) to avoid log(U) = -inf. The CPU sampling algorithm depends on the order of the logits which is undesirable and can also oversample the first logit if it is smaller than the smallest random float larger than 0 (~1e-7). Switching to double precision internally mitigates these problems, although it doesn't fix them. Slowdown is ~35% in the worst case. Also adds various tests that we would like the sampling to pass. CPU Benchmark before: 32 10000 1 0.060 0.069 0.87 32 10000 4 0.229 0.074 3.10 32 10000 32 2.180 0.059 37.09 32 100000 1 0.430 0.480 0.90 32 100000 4 2.322 0.449 5.17 32 100000 32 31.508 0.471 66.96 128 10000 1 0.168 0.235 0.71 128 10000 4 0.965 0.246 3.93 128 10000 32 7.989 0.225 35.51 128 100000 1 1.681 1.539 1.09 128 100000 4 9.012 1.57 35.73 128 100000 32 126.222 1.626 77.60 CPU Benchmark after: 32 10000 1 0.054 0.112 0.48 32 10000 4 0.206 0.093 2.21 32 10000 32 1.826 0.091 20.12 32 100000 1 0.292 0.636 0.46 32 100000 4 2.086 0.606 3.44 32 100000 32 28.496 0.633 45.03 128 10000 1 0.125 0.266 0.47 128 10000 4 0.759 0.258 2.94 128 10000 32 7.362 0.254 29.03 128 100000 1 1.550 2.18 10.71 128 100000 4 8.712 2.22 23.92 128 100000 32 122.585 2.213 55.39 PiperOrigin-RevId: 156917266
* Change the header generated for selective registration to use the isequalsGravatar A. Unique TensorFlower2017-05-23
| | | | | | | defined in the header, instead of strcmp. strcmp is not a constexpr on all platforms. PiperOrigin-RevId: 156914421
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156913869
* Fixes gradient for complex valued inputsGravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156909830
* Fix framework/ops_test.py for Python 3.Gravatar Derek Murray2017-05-23
| | | | PiperOrigin-RevId: 156908714
* Allow complex valued inputs for matrix inverse.Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156908010
* Internal changeGravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156907935
* Register GPU kernels for complex valued inputsGravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156907723
* Add a python test for LayoutOptimizer.Gravatar Yao Zhang2017-05-23
| | | | PiperOrigin-RevId: 156906671
* Clean up some cruft related to d3v4 migration.Gravatar Dandelion Man?2017-05-23
| | | | PiperOrigin-RevId: 156901957
* Bugfix to error message reporting when portpicker is not available.Gravatar Eugene Brevdo2017-05-23
| | | | PiperOrigin-RevId: 156895715
* Bugfixes to tfprof model analyzer unit tests.Gravatar Eugene Brevdo2017-05-23
| | | | PiperOrigin-RevId: 156895551
* Move partial run tests into new PartialRunTest classGravatar Skye Wanderman-Milne2017-05-23
| | | | PiperOrigin-RevId: 156879382
* Add more test cases.Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156876795
* Make nest submodule visible under contrib/framework.Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156875936
* Always return a list for Graph.collection (including in Python3)Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156870745
* Make a silencing context manager for deprecations and use it.Gravatar Martin Wicke2017-05-23
| | | | PiperOrigin-RevId: 156870702
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156870444
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-05-23
| | | | PiperOrigin-RevId: 156869789
* Fill in documentation for SparseFillEmptyRows fastpath c++ opGravatar Eugene Brevdo2017-05-23
| | | | | | Tested by enabling the fast path by default and ensuring unit tests still pass. PiperOrigin-RevId: 156866900
* [XLA] Remove unused InstructionRequiresInputLayoutEqualToOutputLayout hookGravatar Benjamin Kramer2017-05-23
| | | | PiperOrigin-RevId: 156843392
* [XLA] Teach Executable to do its own profiling (patch 3/4).Gravatar A. Unique TensorFlower2017-05-23
| | | | | | | | This completes the sequence by moving ExecuteOnStreamWrapper from xla::Service onto xla::Executable, where it conceptually belongs. The fact that it belongs on Executable is illustrated by the fact that it doesn't access any elements of the Service object, but (in the original form) it takes an Executable object as its first parameter. We can trivially replace this first parameter with an implicit "this". This also removes the (now-unused) "backend" parameter from ExecuteOnStreamWrapper, and moves DumpExecutedHlo from xla::Service to xla::Executable. As with ExecuteOnStreamWrapper, DumpExecutedHlo has no conceptual ties to xla::Service, and the name suggests a clear fit on xla::Executable. PiperOrigin-RevId: 156834459
* Made it possible to refine the behavior of the default op_level_cost_estimator.Gravatar Benoit Steiner2017-05-22
| | | | PiperOrigin-RevId: 156827000
* [XLA] Teach Executable to do its own profiling (patch 2/4).Gravatar A. Unique TensorFlower2017-05-22
| | | | | | | | This follows patch 1 by removing the ShapeSizeBytes function from xla::Compiler, and changing its callers to use Executable::ShapeSizeBytesFunction. Many of the callers are reproducing the same pattern to make a lambda to computes the size of a buffer, and so we condense this into xla::Compiler::BufferSizeBytesFunction. PiperOrigin-RevId: 156814087
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-09 22:08:01 +0000