aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Improve the error message for live set memory check.Gravatar Jianwei Xie2017-05-29
| | | | PiperOrigin-RevId: 157415647
* [tf contrib seq2seq] Bugfixes to BeamSearchDecoderGravatar Eugene Brevdo2017-05-29
| | | | | | | | | | | | | | Implementation by Cinjon Resnick. He can't push this since he's traveling. I just copied the fix and added some small syntax tweaks to make the unit tests pass. More comprehensive unit tests will come in the near future. Fixes at least part of #9904. BeamSearchDecoder: 1. Fix the bug where we don't pass the next cell state through. 2. Gather the cell state (and attention if that's a part of the model as an AttentionWrapper on the cell) according to the next_beam_ids. PiperOrigin-RevId: 157415564
* Fix oversampling in the GPU version of multinomial due to an error in generatingGravatar A. Unique TensorFlower2017-05-29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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: 157414849
* Add numpy_input_fn integration for LinearRegressor and fix the expand_dim ↵Gravatar Jianwei Xie2017-05-29
| | | | | | for label and weight. PiperOrigin-RevId: 157405237
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-05-29
| | | | PiperOrigin-RevId: 157402364
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-05-29
| | | | PiperOrigin-RevId: 157402063
* Fix comment describing ignore_longer_outputs_than_inputs.Gravatar A. Unique TensorFlower2017-05-29
| | | | PiperOrigin-RevId: 157400110
* An initial step of eliminating all implicit broadcast at the HLO level.Gravatar A. Unique TensorFlower2017-05-29
| | | | | | Guard the shape inference for binary ops behind a flag. PiperOrigin-RevId: 157373647
* Set winograd nofused flag to be true by default.Gravatar Yangzihao Wang2017-05-28
| | | | | | Disable winograd nonfused conv for certain input params to avoid a known bug in cuDNNv5 and cuDNNv6. PiperOrigin-RevId: 157352847
* In the CUDA path of depthwise_conv2d, add a fast variant for forward ↵Gravatar A. Unique TensorFlower2017-05-28
| | | | | | convolution when the input images are smaller than 16x16. PiperOrigin-RevId: 157347823
* Fix incorrect condition to instantiate depthwise_ops introduced in commit ↵Gravatar A. Unique TensorFlower2017-05-28
| | | | | | 15d9f00fa. The change should have excluded depthwise_conv2d for doubles on windows debug builds, but it excluded it for all windows and all debug builds. PiperOrigin-RevId: 157345929
* add closing code quotesGravatar Mark Daoust2017-05-28
| | | | PiperOrigin-RevId: 157339360
* Fix the contrib estimator_test by updating the global step in all the ↵Gravatar Jonathan Hseu2017-05-27
| | | | | | appropriate spots. PiperOrigin-RevId: 157328239
* Java: Update Maven release to 1.2.0-rc1Gravatar Asim Shankar2017-05-26
| | | | PiperOrigin-RevId: 157294719
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157292254
* VLOG(2) instead of VLOG(1) for detailed op printouts.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157291238
* tfdbg: add runtime shape and dtype info to DebugNumericSummaryGravatar Shanqing Cai2017-05-26
| | | | PiperOrigin-RevId: 157291215
* Add GraphOptimizer to Grappler item builder to do L1 optimizations andGravatar A. Unique TensorFlower2017-05-26
| | | | | | | | | | | | | | | | | | | | | | | | | | inlining. Op Counts Comparison (BNMT) Counts: Profile vs Grappler Op: Add, 968 vs 965 Op: AddN, 2228 vs 2228 Op: ApplyGradientDescent, 84 vs 84 Op: BatchMatMul, 998 vs 998 Op: Identity, 142 vs 105 Op: MatMul, 63 vs 63 Op: Mul, 10318 vs 10306 Op: OneHot, 1 vs 1 Op: Reshape, 8421 vs 8422 Op: Select, 488 vs 488 Op: Shape, 8132 vs 8131 Op: Sigmoid, 942 vs 942 Op: Softmax, 19 vs 19 Op: StridedSlice, 58 vs 74 Op: Sub, 1398 vs 1394 Op: Tanh, 333 vs 333 Op: Tile, 21 vs 21 Op: Transpose, 39 vs 39 PiperOrigin-RevId: 157288420
* Automated g4 rollback of changelist 157272843Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157288399
* Shift TensorBoard to depend on TensorFlow via `import tensorflow as tf`.Gravatar Dandelion Man?2017-05-26
| | | | | | | | | | | | | | | | | | | | This is in preparation for TensorBoard leaving github/tensorflow/tensorflow and moving into its own repository. There is a little trickiness around tf.contrib.tensorboard, which has the potential to generate circular dependencies because it depends on tensorboard.plugins, which now depends on tensorflow, which depends on contrib. I disconnected the only circle at tensorboard.backend.json_util. Eventually, I hope for tensorboard to have no dependencies on //tensorflow except for files within tensorboard, and on the tensorflow_py target. We need to port or make public a few utility functions (e.g. tensor_util.MakeNdarray) and solve the contrib circular dep before that will be the case. PiperOrigin-RevId: 157288184
* Document layers.l1_l2_regularizer in module docstring so it can be used.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157285789
* Annotate overriding functions with "override" or "final" (and not with ↵Gravatar A. Unique TensorFlower2017-05-26
| | | | | | "virtual") PiperOrigin-RevId: 157284709
* Disable flaky session_test.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157282438
* Don't run failing AssignOpTests with TSAN.Gravatar Skye Wanderman-Milne2017-05-26
| | | | PiperOrigin-RevId: 157281164
* Fix benign "bug" in PyArrayToTFTensorGravatar Skye Wanderman-Milne2017-05-26
| | | | | | | | Passing 'array' to the deallocator didn't do anything since the deallocator doesn't use 'arg', but it's confusing since 'array' is freed by the time the deallocator is called. PiperOrigin-RevId: 157280226
* Extract the histograms dashboard to a proper pluginGravatar A. Unique TensorFlower2017-05-26
| | | | | | | | This continues the great plugin migration. The histograms plugin is very similar to the scalars plugin, except that it only emits JSON and not CSV, so is even simpler. PiperOrigin-RevId: 157279126
* AdagradDA with global step to work with tf.learn models.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157278849
* Fix StopAtStepHook with num_steps when multiple steps are executed in a singleGravatar Jonathan Hseu2017-05-26
| | | | | | session.run(). PiperOrigin-RevId: 157277945
* Fix hash URLsGravatar A. Unique TensorFlower2017-05-26
| | | | | | | | | | | | | | | | | | | | URLs like localhost:6006/#graphs weren't working. This was because the `USE_HASH` property of `TF.Globals` was being read as false by the router, even though `tf-tensorboard` was trying to set it to `true`. The problem was that `tf-tensorboard` was only setting to true the _copy_ of `USE_HASH` that it created at import time. The solution adopted here is to let `globals.ts` close over a private `use_hash`, and expose a getter and setter. An alternate solution is to expose a constant object `{use_hash: false}`; importing modules would copy the _reference_ to this object, so their mutations would pass through. This alternate solution would perhaps be more elegant if we had more properties. Test Plan: - Load http://localhost:6006/. - Change the URL to http://localhost:6006/#graphs. - Note that the view updates. PiperOrigin-RevId: 157277248
* quantization_utils_test is failing on ASAN.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157276855
* Remove unused using-declarationsGravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157276276
* Create a set of sample data for the scalars pluginGravatar A. Unique TensorFlower2017-05-26
| | | | | | | | | | | | | | | | | | | | | | | | | | We aim to provide for each plugin a set of sample data satisfying the following properties: - The data is of a realistic form. - The data is sufficient to exercise all capabilities of the plugin. - The data is not completely boring. Such data then serves many purposes: it is documentation for this particular plugin; it is documentation for TensorFlow and TensorBoard in general; it can be used for manual testing of TensorBoard; and it can be used in the plugin's unit tests. Ideally, we'll also be able to create a script to aggregate the test data from all first-party plugins, providing a single data set that can be used to comprehensively test TensorBoard. To start us off, this change adds such data for the scalars plugin. It's realistic in that it has a few thousand steps and a few dozen runs. It enables testing smoothing (due to the added noise), tag groups (due to the various summary names), and run filters (due to the combinatorial run parameters). And it's implementing a tiny physics simulation, so it's not _entirely_ boring! PiperOrigin-RevId: 157274669
* [XLA] Allow of the tuple instruction as the fusion root.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157274264
* rnn_cell_test is failing on TSAN.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157272979
* Internal changeGravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157272843
* Optionally use new session API in Python APIGravatar Skye Wanderman-Milne2017-05-26
| | | | | | The new session API is enabled if ops._USE_C_API is True, otherwise the original behavior is maintained (the default). Most of the tests in session_test.py are now run with the new API. The new session API runs the TF_Graph that is built alongside the Python graph when ops._USE_C_API is True. PiperOrigin-RevId: 157272139
* Catch more variants of transpose that are simply reshape.Gravatar A. Unique TensorFlower2017-05-26
| | | | | | In particular, transpose and reshape are identical as long as the non-singleton dimensions remain in ascending order in the permutation. PiperOrigin-RevId: 157272046
* Move layout-forcing methods onto HloTestBase where all HLO tests can use themGravatar Eli Bendersky2017-05-26
| | | | PiperOrigin-RevId: 157271246
* Fixes compatibility of contrib feature_column with core make_parse_example_spec.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157270783
* gru_test broken in TSAN.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157270228
* Extend filename sanitizer to cover '[' and ']'.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157269890
* Fix protobuf_util build failure in open source.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157265680
* Fix documentation on weights of tf.losses.softmax_cross_entropy.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157263047
* Fix sign errors in tf.contrib.distributions.logistic.{log_cdf, ↵Gravatar Joshua V. Dillon2017-05-26
| | | | | | log_survival_function} and add accompanying tests. Fixes #10131 PiperOrigin-RevId: 157259406
* Use "override" and "nullptr"; remove unused includesGravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157258631
* Add more information to error message in contrib data function capturing.Gravatar Vijay Vasudevan2017-05-26
| | | | | | | Example error message now looks like: ValueError: Cannot capture a stateful node (name:random_uniform, type:RandomUniformInt) by value. PiperOrigin-RevId: 157258472
* Force GPU device objects that refer to the same physical card using theGravatar A. Unique TensorFlower2017-05-26
| | | | | | | | | same stream id to use the same cuda stream objects. This avoids confusing the per-device memory allocator in ways that cause memory corruption. Fixes https://github.com/tensorflow/serving/issues/335. PiperOrigin-RevId: 157258318
* Automated g4 rollback of changelist 157174708Gravatar Kay Zhu2017-05-26
| | | | PiperOrigin-RevId: 157253080
* Fix "unitialized" typo.Gravatar A. Unique TensorFlower2017-05-26
| | | | PiperOrigin-RevId: 157248921
* Added vocabulary support to canned Estimator multi-class-head.Gravatar Mustafa Ispir2017-05-26
| | | | | | Simplified tests. PiperOrigin-RevId: 157245165
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-08 09:23:47 +0000