aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
...
* [XLA] Update LLVM to upstream revision r304927.Gravatar Peter Hawkins2017-06-07
| | | | | | Add LLVM build rules for the LLVM AMDGPU backend, commented out by default. Fixes issue #10437. PiperOrigin-RevId: 158351480
* Add evaluation test for linear classifier (n==2 or n >2).Gravatar Jianwei Xie2017-06-07
| | | | PiperOrigin-RevId: 158340296
* Set a default for datasets end_of_sequence.Gravatar Brennan Saeta2017-06-07
| | | | | | | | | While all datasets carefully set the end_of_sequence to true at the appropriate time, some datasets might forget to set it to false in the normal case. In order to avoid potential undefined behavior, we set the end_of_sequence variable to be false by default. PiperOrigin-RevId: 158337799
* Setup the env to since ops such as MatchFileOp rely on it.Gravatar Benoit Steiner2017-06-07
| | | | PiperOrigin-RevId: 158336344
* Fix up vz_projector script structureGravatar Justine Tunney2017-06-07
| | | | | | | | We now make sure scripts and HTML imports are declared in the correct places. In the future, pedantically listing script tags should not be necessary. PiperOrigin-RevId: 158334306
* Resubmit a reverted change. Original description:Gravatar Kay Zhu2017-06-07
| | | | | | | | | | | | | | | | | | | | | | [XLA] Enable HloEvaluator for constant folding, also merged a few operations from hlo_constant_folding to hlo_evaluator. Additionally: - In ShapeUtil::ForEachIndex: * fix a bug where visitor is called when the shape has zero elements (e.g., F32{1,0}) * added test case for ForEachIndex. - In HloEvaluator: * Instead of copying and caching a Constant instruction, return the literal directly if the instruction is constant. * Fix an issue where TUPLE and OPAQUE primitives are not keyed in the templated typed_visitor. * Use (fixed) LiteralUtil::Populate to populate resulting literal, fixes the preexisting bug in the evaluator where R0 and shape with zero size dimensions are not handled. * Refactor ElementWiseUnaryOp and HandleCompare to be templatized on the operand's type. * Refactor IsFinite to be top level since it is only applicable to floats and the return type is always boolean. * Change from std::remainder to std::fmod for kRemainder to be compliant with existing XLA behavior. * Change from std::max and std::min to std::fmax and std::fmin to handle NaNs. * Minor comments fix. PiperOrigin-RevId: 158330052
* tf.layers.conv2d use_bias=True to use nn.bias_addGravatar Toby Boyd2017-06-07
| | | | PiperOrigin-RevId: 158326493
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158325855
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158325293
* Add a util function in virtual placer to return canonicalized device string, ↵Gravatar Yuefeng Zhou2017-06-07
| | | | | | which can be used to fix the node's device field before passing them to the maxcut algorithm. PiperOrigin-RevId: 158322753
* [TF:XLA] Initial implementation of TensorArray ops.Gravatar Peter Hawkins2017-06-07
| | | | | | | | | | | The XLA implementation of TensorArrays is more restrictive than regular TensorArrays: * XLA TensorArrays must have dynamic_size=False. * all elements in an XLA TensorArray must have the same shape. * writes always add their values to any existing values; neither reads nor writes ever issue errors. Out-of-bounds writes currently wrap. Refactor Variable handling in the TF/XLA bridge. Use a XlaVariable* to refer to variables inside compilation rather than a numerical ID. Allow for variables that don't correspond to variables known to the user. Also use XlaVariable to handle TensorArrays. PiperOrigin-RevId: 158322041
* [TF:XLA] Refactor randomized tests to allow testing of larger inputs without ↵Gravatar Peter Hawkins2017-06-07
| | | | | | running out of memory. PiperOrigin-RevId: 158321431
* [XLA] Disable constant_folding in test base, so that intended test code pathsGravatar Kay Zhu2017-06-07
| | | | | | would not be elided by constant_folding pass. PiperOrigin-RevId: 158317641
* Fix typo.Gravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158310742
* Swap the order of NanTensorHook and custom hooksGravatar Shanqing Cai2017-06-07
| | | | | | to ensure that when the training encounteres NaN's in the loss function, user-supplied hooks such as tf_debug.LocalCLIDebugHook can still be used to debug the root cause of the numeric issues. PiperOrigin-RevId: 158310249
* [XLA] Propagate debug option flags to hlo_test_base.Gravatar Eli Bendersky2017-06-07
| | | | | | | | | | | | | Specific HLO tests have to replace the generic test_main target with a manual main() that invokes RUN_ALL_TESTS. To get access to a module with debug options set up, a new convenience method is created on HloTestBase. Initially algebraic_simplifier_test is modified as a canary; in a followup we'll convert all HLO tests to this approach. PiperOrigin-RevId: 158309488
* [Tensorboard] Add a trace viewer component to TensorBoard.Gravatar Eric Liu2017-06-07
| | | | | | | | | | | | | | | | | We make the trace viewer a separate app; otherwise, there would be dependency conflicts (e.g. Polymer) between the trace viewer app and the tensorboard app. The trace viewer app would be served by a plugin, and Tensorboard dashboard will integrate trace viewer app using iframe in the future. This CL also added "mominify" support for link import HTML tags in the tensorboard home-grown java vulnizer; otherwise, the vulcanized trace viewer code would crash the java vulcanizer. For open-source build, we add a denpendency on the Catapult github repository (https://github.com/catapult-project/catapult/tree/master/tracing). We use a bazel genrule to vulcanize a trace viewer binary which is then used in the tf-trace-viewer component. PiperOrigin-RevId: 158309408
* Support unknown emit shapes in tf.nn.raw_rnn.Gravatar RJ Ryan2017-06-07
| | | | PiperOrigin-RevId: 158308002
* Add label-vocab support to binary logistic head.Gravatar Mustafa Ispir2017-06-07
| | | | | | | Add assertion that binary classifier label is in range [0., 1.] Fixed Classifier Integration tests. PiperOrigin-RevId: 158307521
* In the CUDA path of depthwise_conv2d, optimize backward filter convolution ↵Gravatar A. Unique TensorFlower2017-06-07
| | | | | | for images 2 or 4 times smaller than 16x16. Also initialize in_cols from blockDim, to fix the regression caused in CL 157906773. PiperOrigin-RevId: 158296136
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158295169
* Add bazel version check to .configureGravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158294569
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158294289
* Replace std::function object with regular function.Gravatar A. Unique TensorFlower2017-06-07
| | | | | | The function is called recursively, and the std::function object had only existed to allow recursion from within a lambda expression. A regular function should be cheaper than a polymorphic function wrapper. PiperOrigin-RevId: 158292415
* Use template specialization instead of overloaded methods. This is a more ↵Gravatar A. Unique TensorFlower2017-06-07
| | | | | | appropriate tool here. NFC PiperOrigin-RevId: 158292035
* Make tf.contrib.lookup python functions use the kernels v2 that uses the ↵Gravatar Yutaka Leon2017-06-07
| | | | | | resource tensor as handler. PiperOrigin-RevId: 158291836
* Switch back to max_num_rows_to_load instead of reading slice by slice due to ↵Gravatar Wei Ho2017-06-07
| | | | | | | | performance regression from network overhead. Add check when using initializing values to avoid seg fault PiperOrigin-RevId: 158291218
* Support numpy-style padding and slicing of tf.spectral.rfft/irfft to match ↵Gravatar RJ Ryan2017-06-07
| | | | | | | | the desired FFT length. Fixes incorrect RFFT/IRFFT results when fft_length does not match the input dimension. PiperOrigin-RevId: 158289991
* Update iOS examples to use CocoaPods, and moved to tensorflow/examples/iosGravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158289285
* Clean up some code after previous CLGravatar Eli Bendersky2017-06-07
| | | | PiperOrigin-RevId: 158282834
* Adds ability to set a "family" attribute in Tensorflow summaries, whichGravatar A. Unique TensorFlower2017-06-07
| | | | | | | | controls the "tab name" of the summary that is displayed. This solution keeps using name_scope to keep names unique, but then prefixes the tag with the family name if provided. PiperOrigin-RevId: 158278922
* Adds integration test for DNNLinearCombined((Classifier)|(Regressor)).Gravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158278512
* Remove a further unused proto header inclusionGravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158278026
* [XLA] Add HloLocation to dataflow analysis.Gravatar Mark Heffernan2017-06-07
| | | | | | | | Add an HloLocation abstraction to dataflow analysis which indicates where (in the output of what instruction and at which index) an HloValue may appear. Previously only uses were stored with an HLO value where a use is an edge in the HLO graph (instruction, operand number and ShapeIndex). Also, change the handling of tuple-shaped kSelect instructions when ssa_form is true. Previously a phi value would be created. With this change the the value set instead contains the union of it's inputs identical to the ssa_form=false case. PiperOrigin-RevId: 158276598
* [XLA] Start collecting flags for debug options in a single place.Gravatar Eli Bendersky2017-06-07
| | | | | | | | | | | | | | | ClientLibraryTestBase will now parse command-line flags for debug options automatically, permitting subclasses to override certain options by using mutable_debug_options. main() still has to call AppendDebugOptionsFlags() explicitly before running the TF flag parser. In the mean-time, this CL leaves flag handling to the current "legacy" approach. However, this is part of a larger plan to move *all* debugging flags for XLA into the DebugOptions message and expose them as flags from a single place. The other flags (which are not controlling debugging options) will have to be propagated more explicitly. PiperOrigin-RevId: 158276294
* Properly handle shape nodes that have a preexisting control dependencyGravatar Benoit Steiner2017-06-07
| | | | PiperOrigin-RevId: 158274845
* Minor cleanupGravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158268933
* Sort header inclusions; define EIGEN_USE_THREADS where headers depend on it.Gravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158267803
* Add missing header inclusionGravatar A. Unique TensorFlower2017-06-07
| | | | PiperOrigin-RevId: 158265934
* When GPU is configured, do not require --config=cuda.Gravatar Gunhan Gulsoy2017-06-06
| | | | | | Also fix indentation in configure. PiperOrigin-RevId: 158232959
* Java: Update Maven release to 1.2.0-rc2Gravatar Asim Shankar2017-06-06
| | | | PiperOrigin-RevId: 158212897
* Profile memory usage in VirtualScheduler and report peak memory usage.Gravatar A. Unique TensorFlower2017-06-06
| | | | | | | | | To do so, NodeState now handles different output ports of a node (in case a node has multiple outputs). Also, VirtualScheduler code is cleaned up with more comments. PiperOrigin-RevId: 158209068
* Add a frontend for viewing the first ops that exhibit bad values (NaN, +/- Inf).Gravatar A. Unique TensorFlower2017-06-06
| | | | | | This helps the user identify problematic ops. Also moved the debugger data logic within tf-graph-info into a new tf-graph-debugger-data-card component. PiperOrigin-RevId: 158208679
* Making load() work for resource variables.Gravatar Alexandre Passos2017-06-06
| | | | PiperOrigin-RevId: 158205361
* [XLA] Simplify Shape traversal visitors.Gravatar Mark Heffernan2017-06-06
| | | | | | Simplify shape traversal visitors in ShapeUtil and ShapeTree. Add a non-Status form because most uses of the traversal methods do not use it, and remove is_leaf parameter from ShapeTree.ForEach* as it is not frequently used. PiperOrigin-RevId: 158201574
* Extracted linear estimator testing utils to be reused by dnn-linear-combined.Gravatar Mustafa Ispir2017-06-06
| | | | | | Added tests for linear part of dnn-linear-combined estimator. PiperOrigin-RevId: 158200827
* Add arrowheads to dataflow edges.Gravatar A. Unique TensorFlower2017-06-06
| | | | | | | | | | | Make reference edges orange. Remove animations from tooltips in the graph documentation. Previously, arrowheads were only added to reference edges (because we assumed users knew about the convention that arrowless edges flow upwards). That decision nicely reduces clutter. However, recently, some internal and external folks have expressed confusion, and so I want to try adding arrowheads to all data flow edges. And make the reference edges starkly different. See #10428 PiperOrigin-RevId: 158195388
* Add only string constants to ASSET_FILEPATHS collection.Gravatar David Soergel2017-06-06
| | | | PiperOrigin-RevId: 158192152
* Add tests with different delta to huber_loss.Gravatar Sergio Guadarrama2017-06-06
| | | | PiperOrigin-RevId: 158191361
* Fixes a bug in setting default optimizers for DNNLinearCombinedClassifier.Gravatar A. Unique TensorFlower2017-06-06
| | | | PiperOrigin-RevId: 158190192
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-03 13:28:36 +0000