aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Remove name_scope from convolutional calls.Gravatar A. Unique TensorFlower2017-10-23
| | | | PiperOrigin-RevId: 173171871
* Makes gradients_function exception-safe.Gravatar Alexandre Passos2017-10-23
| | | | PiperOrigin-RevId: 173170394
* Remove contrib/xla_tf_graph/Gravatar Sanjoy Das2017-10-23
| | | | | | It isn't needed anymore. PiperOrigin-RevId: 173165310
* Functions in graph mode can call other functions in graph mode.Gravatar Alexandre Passos2017-10-23
| | | | PiperOrigin-RevId: 173163050
* Rename add_layer -> track_layer in Network.Gravatar A. Unique TensorFlower2017-10-23
| | | | PiperOrigin-RevId: 173159032
* Automated g4 rollback of changelist 172936802Gravatar A. Unique TensorFlower2017-10-23
| | | | PiperOrigin-RevId: 173158990
* [TF:XLA] Replace a Mul in XlaArgMinMaxOp with an AndGravatar A. Unique TensorFlower2017-10-23
| | | | PiperOrigin-RevId: 173155697
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-10-23
| | | | PiperOrigin-RevId: 173147086
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-10-23
| | | | PiperOrigin-RevId: 173145770
* Cut tracing memory costGravatar A. Unique TensorFlower2017-10-23
| | | | PiperOrigin-RevId: 173144626
* [XLA] Erase cloned instructions on the fly when merging fusion nodes.Gravatar A. Unique TensorFlower2017-10-23
| | | | | | This avoids the awkward situation where an RNG which is clearly eligible for fusion becomes ineligible mid-fusion because it suddenly has an extra (dead) user. PiperOrigin-RevId: 173141716
* Add SerializeIterator op that serializes an IteratorResource into a variant ↵Gravatar Saurabh Saxena2017-10-23
| | | | | | | | | | | tensor. Add DeserializeIterator op that builds IteratorResource from a variant tensor. Move BundleReaderWrapper and BundleWriterWrapper from dataset.h to iterator_ops.cc. Add generic key-value store interfaces IteratorStateReader and IteratorStateWriter for reading/writing state of iterators. Get rid of IteratorBundleReader and IteratorBundleWriter. PiperOrigin-RevId: 173140858
* Internal changeGravatar A. Unique TensorFlower2017-10-23
| | | | PiperOrigin-RevId: 173136642
* Fix breakages in OSS buildsGravatar Shanqing Cai2017-10-23
| | | | | | | | | | | | | See example breakages logs at: http://ci.tensorflow.org/job/tensorflow-cl-cpu-python3-pip/10847/console http://ci.tensorflow.org/job/tensorflow-cl-gpu/11008/console 1. CL/172477381 added the no_oss tag to tests with oss_serial tags, which broke the logic of OSS_SERIAL tests in pip.sh and run_pip_test.sh. This CL fixes that. 2. The nccl_kernels BUILD target in contrib/nccl/BUILD was missing some dependencies. This CL adds the missing ones. Fixes: #13918 PiperOrigin-RevId: 173133914
* Allows calling keras layers in eager mode.Gravatar Alexandre Passos2017-10-23
| | | | PiperOrigin-RevId: 173129805
* Switching contrib.summaries API to be context-manager-centricGravatar Alexandre Passos2017-10-23
| | | | PiperOrigin-RevId: 173129793
* Put Bazel mirror URLs firstGravatar Justine Tunney2017-10-23
| | | | PiperOrigin-RevId: 173127955
* [XLA] Add support for convolutions with no spatial dimensionsGravatar David Majnemer2017-10-23
| | | | PiperOrigin-RevId: 173126950
* [tf.data] Convert dataset arguments to tensors as early as possible.Gravatar Derek Murray2017-10-23
| | | | | | | This change raises a `TypeError` earlier if (for example) the `batch_size` argument to `Dataset.batch()` has the incorrect type. PiperOrigin-RevId: 173126678
* K-FAC: Support for registering multiple minibatches with ↵Gravatar A. Unique TensorFlower2017-10-23
| | | | | | register_fully_connected() PiperOrigin-RevId: 173121735
* Avoid listing all modified Enter/RefEnter nodes on INFO, use VLOG(1) instead.Gravatar Tim Harley2017-10-23
| | | | | | Leave a single, simple, message on INFO. PiperOrigin-RevId: 173121726
* K-FAC: _check_registration() supports multiple towers.Gravatar A. Unique TensorFlower2017-10-23
| | | | PiperOrigin-RevId: 173115870
* Multi-minibatch support forGravatar A. Unique TensorFlower2017-10-23
| | | | | | tf.contrib.kfac.fisher_blocks.FullyConnectedKFACBasicFB. PiperOrigin-RevId: 173109677
* Fix import of meta graphs with partitioned variables into a scope.Gravatar A. Unique TensorFlower2017-10-23
| | | | | | | | | | | | Saver inspects SliceInfo to decide the variable name when creating a checkpoint. Before this fix even if a partitioned variable ("weights") was imported into a scope "a" it would still be checkpointed as ("weights") instead of ("a/weights") since import_scoped_meta_graph was not adjusting the SliceInfo. WARNING: if you use import_meta_graph on graphs with partitioned_variables WITH an import_scope argument AND then create a Saver to write/read checkpoints this change may break your checkpoint loading. PiperOrigin-RevId: 173105796
* K-FAC: Multi-tower support for ConvDiagonalFB.Gravatar A. Unique TensorFlower2017-10-23
| | | | PiperOrigin-RevId: 173105412
* Add tests for convolution 1DGravatar Yunxing Dai2017-10-22
| | | | | | RELNOTES: n/a PiperOrigin-RevId: 173060283
* Update pin for bazel-toolchains to latest versionGravatar A. Unique TensorFlower2017-10-21
| | | | PiperOrigin-RevId: 173002530
* Slight change to reduce_test to avoid generating inf, which was triggering ↵Gravatar Bjarke Hammersholt Roune2017-10-20
| | | | | | an inf detector unnecessarily. PiperOrigin-RevId: 172965466
* Adding Python ApiDef overrides.Gravatar Anna R2017-10-20
| | | | PiperOrigin-RevId: 172960496
* Internal change.Gravatar Anna R2017-10-20
| | | | PiperOrigin-RevId: 172960439
* Add dtype argument to Mean and Accuracy object-oriented metrics.Gravatar A. Unique TensorFlower2017-10-20
| | | | PiperOrigin-RevId: 172957714
* Adding the Stanford Tensorflow class to community resources.Gravatar A. Unique TensorFlower2017-10-20
| | | | PiperOrigin-RevId: 172956049
* Removed some unnecessary broadcasts in binary ops where only one input needsGravatar A. Unique TensorFlower2017-10-20
| | | | | | broadcasting (which is a fairly common case, even in the fallback path). PiperOrigin-RevId: 172950493
* Make `tf.contrib.distributions` quadrature family accept a `Tensor` forGravatar Joshua V. Dillon2017-10-20
| | | | | | `quadrature_grid_and_probs` argument. PiperOrigin-RevId: 172950094
* Dump the computation's SessionModule as part of the tf_compile rule.Gravatar Justin Lebar2017-10-20
| | | | PiperOrigin-RevId: 172946149
* Add streaming_precision_recall_at_equal_thresholdsGravatar A. Unique TensorFlower2017-10-20
| | | | | | | | This helper method computes streaming tp, fp, tn, fp, precision, and recall for the user in a way that exhibits O(T + N) time and space complexity (instead of O(T * N)), where T is the number of thresholds and N is the size of the predictions tensor. Thanks to Frank Chu for the efficient algorithm! PiperOrigin-RevId: 172946073
* Log Hlo IR during AOT compilationGravatar Sanjoy Das2017-10-20
| | | | PiperOrigin-RevId: 172944165
* Allows tfe.enable_eager_execution(device_policy=tfe.DEVICE_POLICY_WARN).Gravatar Alexandre Passos2017-10-20
| | | | PiperOrigin-RevId: 172943398
* Merge branch 'master' of sso://tensorflow/stagingGravatar TensorFlower Gardener2017-10-20
|\
| * Add performance guide for fused decode_and_crop_jpeg optimization.Gravatar Mingxing Tan2017-10-20
| | | | | | | | PiperOrigin-RevId: 172943116
| * Make Network compatible with eager mode. Currently it only allows to ↵Gravatar Francois Chollet2017-10-20
| | | | | | | | | | | | instantiate a Network in eager mode using the regular Keras API, and call it on eager tensors. PiperOrigin-RevId: 172942569
* | Testing pending CL: 172939383Gravatar ashankar2017-10-20
|/
* Simplifies capturing code in graph_callable to use recent function improvements.Gravatar Alexandre Passos2017-10-20
| | | | PiperOrigin-RevId: 172937003
* Automated g4 rollback of changelist 172924803Gravatar A. Unique TensorFlower2017-10-20
| | | | PiperOrigin-RevId: 172936802
* Add a BUILD file for makefile build.Gravatar Gunhan Gulsoy2017-10-20
| | | | PiperOrigin-RevId: 172933558
* Updating to latest stable version of llvm repoGravatar A. Unique TensorFlower2017-10-20
| | | | | | (same used to generate latest in gcr.io/cloud-marketplace/google/clang-debian8) PiperOrigin-RevId: 172930105
* Simplify the graph generated for contrib/summaries in theGravatar A. Unique TensorFlower2017-10-20
| | | | | | "always summarize" and "never summarize" cases by skipping the `cond`. PiperOrigin-RevId: 172928083
* Add AdaptiveSharedBatchScheduler which processes batches at a variable rate ↵Gravatar A. Unique TensorFlower2017-10-20
| | | | | | which can be adjusted based on external feedback. For reasonable feedback, this scheduler should deliver better latency than the SharedBatchScheduler. PiperOrigin-RevId: 172924803
* Expose tf.contrib.framework.current_arg_scope()Gravatar A. Unique TensorFlower2017-10-20
| | | | PiperOrigin-RevId: 172922818
* Enables silent copies of eager tensors for specially-constructed contexts.Gravatar Alexandre Passos2017-10-20
| | | | PiperOrigin-RevId: 172922467
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-04 05:35:13 +0000