aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Automated rollback of commit 7bbd69fd16d4e97afd417786c1ee2fff27d92703Gravatar Rohan Jain2018-09-24
| | | | PiperOrigin-RevId: 214321627
* Split up SPARSE_DEPS, adding each individual dependency only to the sparseGravatar A. Unique TensorFlower2018-09-24
| | | | | | operators that need it. PiperOrigin-RevId: 214320700
* Make sure broken tests are filtered out in XLA tests suites.Gravatar Gunhan Gulsoy2018-09-24
| | | | PiperOrigin-RevId: 214311663
* [XLA] Update Scatter to ignore out-of-bound indices.Gravatar A. Unique TensorFlower2018-09-24
| | | | PiperOrigin-RevId: 214309598
* Updated to newest FlatBuffers and FlexBuffers.Gravatar A. Unique TensorFlower2018-09-24
| | | | PiperOrigin-RevId: 214309210
* Wrap forward and backward pass in a defun for L2HMC.Gravatar Akshay Modi2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Also a small bugfix to handle unknown shapes in backprop._num_elements. Before: entry { name: "L2hmcBenchmark.eager_train_cpu_defun" iters: 10 wall_time: 0.594115018845 extras { key: "examples_per_sec" value { double_value: 336.635152548 } } } After: entry { name: "L2hmcBenchmark.eager_train_cpu_defun" iters: 10 wall_time: 0.322251081467 extras { key: "examples_per_sec" value { double_value: 620.634069216 } } } PiperOrigin-RevId: 214308142
* Further simplify the cuDNN wrappers. Instead of passing aroundGravatar Tim Shen2018-09-24
| | | | | | | | | | CudnnConvParams, just pass around the HloInstruction. This is based on the observation that most code doesn't care about the convolution semantics like which operand is input vs filter vs output. In fact, only layout assignment and conv runner care about them. PiperOrigin-RevId: 214307399
* Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-09-24
| | | | | | | | self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about: * the fact that the session may be reused. * the session is not closed even when doing a "with self.test_session()" statement. PiperOrigin-RevId: 214300210
* This CL adds an init_scope to the Keras set & get learning phase functions. ↵Gravatar A. Unique TensorFlower2018-09-24
| | | | | | | This allows the Keras learning phase to work inside functions and defuns. Note: There might still be bugs in graph mode if the default placeholder is being fed (instead of using set_learning_phase) and a layer is in a function. PiperOrigin-RevId: 214299002
* [Java]: Release 1.11.0-rc2Gravatar Asim Shankar2018-09-24
| | | | PiperOrigin-RevId: 214298224
* Merge pull request #22350 from yongtang:22295-gradient_multipliersGravatar TensorFlower Gardener2018-09-24
|\ | | | | | | PiperOrigin-RevId: 214297796
* | Disabling MultiDeviceIterator in Eager mode. Support is coming soon.Gravatar Rohan Jain2018-09-24
| | | | | | | | PiperOrigin-RevId: 214296771
* | Add functionality to SubSlice a tensor.Gravatar A. Unique TensorFlower2018-09-24
| | | | | | | | PiperOrigin-RevId: 214295534
* | [data-stats] Exposes `StatsAggregator` and `set_stats_aggregator` in ↵Gravatar Shivani Agrawal2018-09-24
| | | | | | | | | | | | tf.contrib.data. PiperOrigin-RevId: 214294955
* | Fix documentation in some lookup_ops and add eager test.Gravatar Yutaka Leon2018-09-24
| | | | | | | | PiperOrigin-RevId: 214294594
* | Merge pull request #21798 from facaiy:ENH/div_no_nan_treate_negative_as_zeroGravatar TensorFlower Gardener2018-09-24
|\ \ | | | | | | | | | PiperOrigin-RevId: 214290400
* \ \ Merge pull request #22469 from rfernand2:masterGravatar TensorFlower Gardener2018-09-24
|\ \ \ | | | | | | | | | | | | PiperOrigin-RevId: 214289882
* | | | Remove the pretty formatting of generated code. The astor library that did ↵Gravatar Dan Moldovan2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | that uses circular references, which can be problematic in eager mode. PiperOrigin-RevId: 214287432
* | | | Clear the stale README and add reference to the updated file.Gravatar Dan Moldovan2018-09-24
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 214286888
* | | | Replace self.test_session(graph=<an object not None>) with ↵Gravatar A. Unique TensorFlower2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | self.session(graph=...) as it's the same semantic. PiperOrigin-RevId: 214286845
* | | | Implement required properties for TPU StrategyGravatar Philip Pham2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | These properties are necessary for the strategy to work with `tf.estimator.train_and_evaluate`. PiperOrigin-RevId: 214285957
* | | | Tweak PrintInterpreterState() outputGravatar Jared Duke2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Output names for custom ops and properly insert newlines for null IntVectors. PiperOrigin-RevId: 214285696
* | | | Check for too large jpeg images before decompressing.Gravatar Mihai Maruseac2018-09-24
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 214279868
* | | | Merge pull request #18567 from imsheridan:fix_expand_dimsGravatar TensorFlower Gardener2018-09-24
|\ \ \ \ | | | | | | | | | | | | | | | PiperOrigin-RevId: 214278672
* | | | | Remove unused compatibility code in Softmax implementation now that the ↵Gravatar Peter Hawkins2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | forward compatibility window has expired. PiperOrigin-RevId: 214277870
* | | | | Turn on PinToHostOptimizer by default.Gravatar A. Unique TensorFlower2018-09-24
| | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214275960
* | | | | Merge pull request #22461 from feihugis:Issue-12414_Error_Logging_SessionGravatar TensorFlower Gardener2018-09-24
|\ \ \ \ \ | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214271324
* \ \ \ \ \ Merge pull request #22457 from elmsfu:tflite/make_abslGravatar TensorFlower Gardener2018-09-24
|\ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214269484
* | | | | | | Rename source_map to create_source_map. Reorganize the tests to be clearer ↵Gravatar Dan Moldovan2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | about the expected functionality. PiperOrigin-RevId: 214266947
* | | | | | | Deprecate do_quantize_training_on_graphdef.Gravatar Suharsh Sivakumar2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214263489
* | | | | | | Add a rounding operation to eliminate some truncation errors (see test for ↵Gravatar A. Unique TensorFlower2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | example). PiperOrigin-RevId: 214262549
* | | | | | | Non-injective bijectors should still support the fallback from ildj to -fldj ↵Gravatar Brian Patton2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | and vice versa. PiperOrigin-RevId: 214257872
* | | | | | | Move a couple more docs from tf.contrib -> tfpGravatar Brian Patton2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214255941
* | | | | | | Add support for non-string attributesGravatar A. Unique TensorFlower2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214251264
* | | | | | | Generalize sort implementation in the HloEvaluator.Gravatar Adrian Kuegel2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It only worked for ranks 1 or 2, and only if the dimension to sort is the most minor dimension. Also fix the SafeLess function so that the SortExtremeValues() test passes. PiperOrigin-RevId: 214239560
* | | | | | | Clean-up of function.py.Gravatar Lasse Espeholt2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214232622
* | | | | | | compat: Update forward compatibility horizon to 2018-09-24Gravatar A. Unique TensorFlower2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214230777
* | | | | | | Merge pull request #22415 from yongtang:22372-verbsGravatar TensorFlower Gardener2018-09-24
|\ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214221868
* \ \ \ \ \ \ \ Merge pull request #22060 from perfinion:systemlibsGravatar TensorFlower Gardener2018-09-24
|\ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214220618
* | | | | | | | | Remove dependency on contrib dataset ops.Gravatar Priya Gupta2018-09-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214219282
* | | | | | | | | Merge pull request #22352 from Intel-tensorflow:agramesh/disable_mkl_ml_onlyGravatar TensorFlower Gardener2018-09-23
|\ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214215589
* | | | | | | | | | Fix noop elimination optimization.Gravatar Piotr Padlewski2018-09-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix for b/116169724 Only remove noops if they refer to const nodes. PiperOrigin-RevId: 214199200
* | | | | | | | | | [tf.data] Add `tf.contrib.data.Optional` support to `Structure`.Gravatar Derek Murray2018-09-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change switches `tf.contrib.data.Optional` to use a `Structure` class to represent the structure of its value, instead of `output_types`, `output_shapes`, and `output_classes` properties. It adds support for nesting `Optional` objects and representing their structure. This change also makes a modification to the `Structure` class: `Structure.is_compatible_with(x)` now takes another `Structure` as the `x` argument, instead of a value. This makes it easier to work with nested structures (where we might not have a value readily available), and better matches the interface of other `is_compatible_with()` methods (e.g. in `tf.TensorShape` and `tf.DType`). Finally, in the process of making this change, I observed possible crash-failures when a DT_VARIANT tensor containing another DT_VARIANT tensor is copied between CPU and GPU. This change "fixes" the immediate problem by raising an UnimplementedError, but more work will be necessary to support the full range of use cases. PiperOrigin-RevId: 214198993
* | | | | | | | | | Add blacklist ops to PinToHostOptimizer. Fix test.Gravatar A. Unique TensorFlower2018-09-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214195020
* | | | | | | | | | Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2018-09-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214179792
* | | | | | | | | | Add Metadata object for RecordReader and associated stats computing /Gravatar Vijay Vasudevan2018-09-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | MD fetching method. TFRecord files do not contain a file-level header that describes the MD of the file. To avoid backwards compatibility issues, we add a lightweight function that computes the statistics over the file once and caches the result for future calls. A future implementor could do a better job of computing GetMetadata() by having the RecordWriter emit these entries during writing so that GetMetadata() only reads. Doing so will require additional backwards compatibility checks to ensure that the function works both for old TFRecords and the new format. PiperOrigin-RevId: 214178704
* | | | | | | | | | Merge pull request #21715 from hsgkim:volume_patchesGravatar TensorFlower Gardener2018-09-23
|\ \ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214177065
* | | | | | | | | | | Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2018-09-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214176089
* | | | | | | | | | | Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2018-09-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214174935
* | | | | | | | | | | Moving MultiDeviceIterator from contrib to core.Gravatar Rohan Jain2018-09-23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 214173896
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-01 13:58:44 +0000