aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
...
* Reuses numpy memory for fed Tensors.Gravatar Alexandre Passos2017-03-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Waits until we are holding the GIL to lazily dereference the objects. Session benchmarks, before: benchmark_session_feed_direct_4B 1 0.000151 benchmark_session_feed_direct_4MB 1048576 0.000633 benchmark_session_fetch_direct_4B 1 0.000107 benchmark_session_fetch_direct_4MB 1048576 0.000481 benchmark_session_feed_grpc_4B 1 0.000307 benchmark_session_feed_grpc_4MB 1048576 0.000973 benchmark_session_fetch_grpc_4B 1 0.000192 benchmark_session_fetch_grpc_4MB 1048576 0.000876 After: benchmark_session_feed_direct_4B 1 0.000150 benchmark_session_feed_direct_4MB 1048576 0.000153 benchmark_session_fetch_direct_4B 1 0.000106 benchmark_session_fetch_direct_4MB 1048576 0.000507 benchmark_session_feed_grpc_4B 1 0.000305 benchmark_session_feed_grpc_4MB 1048576 0.000360 benchmark_session_fetch_grpc_4B 1 0.000225 benchmark_session_fetch_grpc_4MB 1048576 0.000873 So feed_direct_4m is ~3x faster and feed_grpc_4m is ~2.5x faster Change: 151170284
* Android demo: prefer specific preview frame sizes if available. Will at ↵Gravatar Andrew Harp2017-03-24
| | | | | | least partially address #7464. Change: 151168988
* Fix tfprof to measure the actual kernel execution time.Gravatar A. Unique TensorFlower2017-03-24
| | | | Change: 151166174
* Strength reduce Dot into broadcasting multiply and reduce. Also optimizesGravatar Blake Hechtman2017-03-24
| | | | | transposes and reshapes that feed reductions. Change: 151162327
* Java: Maven: Fix some boo boos with the Windows installation.Gravatar Asim Shankar2017-03-24
| | | | | | | - The resource path should not use the platform's file separator. - Restore the commands to download and extract the Windows native libraries. - Unrelated tweaks the README Change: 151161421
* - Added accompanying .py file for deep MNIST tutorialGravatar A. Unique TensorFlower2017-03-24
| | | | Change: 151158189
* Fix lint issues after pull.Gravatar Martin Wicke2017-03-24
| | | | Change: 151154030
* Improved transpose operator's performanceGravatar Yangzihao Wang2017-03-24
| | | | | Use specialized GPU kernels on tensors when the permutation can be reduced to {0,2,1}, {2,1,0} or {1,0}. Change: 151147354
* Move NodeMap to utils.ccGravatar Yao Zhang2017-03-24
| | | | Change: 151147266
* Added a config option to control model pruningGravatar Benoit Steiner2017-03-24
| | | | Change: 151130707
* Bring the Load Data button to the projector in tensorboard, just for metadata.Gravatar Nikhil Thorat2017-03-24
| | | | Change: 151128714
* Bump pooling_ops_3d_test size to mediumGravatar Gunhan Gulsoy2017-03-24
| | | | Change: 151103646
* Java: Scripts to package and upload to Maven Central.Gravatar Asim Shankar2017-03-23
| | | | | | | | | | | The general idea is that the scripts here download release artifacts (Java sources, JNI native libraries) built using bazel as part of the TensorFlow release process and packages them into a maven repository. See README.md for instructions and layout of the artifacts. Final step for #6926 Change: 151096674
* Autogenerated Change: Change TensorBoard TAG to 50Gravatar Dandelion Mané2017-03-23
| | | | Change: 151087060
* Update Experiment to support core Estimator.Gravatar Jianwei Xie2017-03-23
| | | | Change: 151086446
* Added a method to grappler_item to return all the variables accessed duringGravatar Benoit Steiner2017-03-23
| | | | | inference or training Change: 151086011
* Internal changesGravatar Rohan Jain2017-03-23
| | | | Change: 151077399
* Update LLVM version to upstream revision r298633.Gravatar Mark Heffernan2017-03-23
| | | | Change: 151077249
* Use gather based SliceProcessor if the input shape is not a const but derivedGravatar Yao Zhang2017-03-23
| | | | | from other consts. Change: 151075063
* Delete C library handle when done to avoid leaking memory.Gravatar Asim Shankar2017-03-23
| | | | Change: 151074297
* Tweaks for selecting y axes for plots, especially ones with small ranges.Gravatar A. Unique TensorFlower2017-03-23
| | | | | | | | | | | | | The previous range selection heuristic tended to be too aggressive in expanding the range of the y-axis when dealing with values close to 0. For example, if all of the values ranged between 0.65 and 0.7 then it would set the y-range to somewhere near [-0.1, 0.7], making it impossible to see changes over time. If all the values [0, 1e-4] then it would select a range close to [-0.1, 1e-4]. This change adds logic to avoid expanding the range of the y-axis too much. It also modifies the 'ignore outliers' logic to select a range that includes 90% of the data, as opposed to the checked-in version that uses 80%. Change: 151071239
* Raise an error for SyncReplicasOptimizer in DNNLinearCombinedGravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151071183
* softmax_loss_function in tf.contrib.legacy_seq2seq and tf.contrib.seq2seq ↵Gravatar Neal Wu2017-03-23
| | | | | | | now requires named arguments in order to prevent mixing up labels and logits Also added lint fixes and fixed a broken/hidden test that wasn't being run Change: 151069880
* tfdbg core: explicitly disable debug_gateway_test on GPU with tagGravatar Shanqing Cai2017-03-23
| | | | Change: 151065650
* Edits and fixes to SDCA-based tf.learn estimators:Gravatar Petros Mol2017-03-23
| | | | | | - Renaming SDCARegressor to SDCALinearRegressor - Improving documentations Change: 151065563
* Internal change.Gravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151064926
* A new guide: "Installing TensorFlow for C"Gravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151064589
* Switch 3d pooling to the unified volume extraction op. Reduce tensor sizes ↵Gravatar A. Unique TensorFlower2017-03-23
| | | | | | in test cases, and switch the size of pooling_ops_3d_test back to "small". Delete the Extract3dPatch function which is not used anywhere anymore. Change: 151062901
* Android: update TensorFlowInferenceInterface: replace error code returns ↵Gravatar Andrew Harp2017-03-23
| | | | | | | with exceptions, use longs for dimensions, and simplify method signatures where possible. Note this is a breaking change and will cause the TF demo in tensorflow/examples/android to not work with older jars/.so binaries built from tensorflow/contrib/android Change: 151061702
* [XLA] Add TruncatedNormal.Gravatar A. Unique TensorFlower2017-03-23
| | | | | Add "Any" to client library routines and add tests for it. Change: 151060097
* Adjust the create_train_op to have a flag for check_numerics.Gravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151050903
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151050198
* Allow users to specify the RNNCell type for the StateSavingRnnEstimator.Gravatar John Bates2017-03-23
| | | | Change: 151049836
* Relax absolute tolerance for conv3d GPU tests, and add a relative tolerance ↵Gravatar A. Unique TensorFlower2017-03-23
| | | | | | threshold. Reenable conv_ops_3d_test. Change: 151049454
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151049375
* Add OnesLike op and kernel.Gravatar Suharsh Sivakumar2017-03-23
| | | | Change: 151048951
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151048554
* Update the docstring for continuous_eval.Gravatar Jianwei Xie2017-03-23
| | | | Change: 151047342
* Merge changes from github.Gravatar Martin Wicke2017-03-23
| | | | Change: 151046259
* Target devices should use tf_cuda_library.Gravatar Yao Zhang2017-03-23
| | | | Change: 151042735
* The "metadata" fields from the internal json.Gravatar Mark Daoust2017-03-23
| | | | | These were producing a warning popup when you run the notebook in the current jupyter (4.3.0). Change: 151041728
* FixedLenSequenceFeature for parsing a variable-length tensor with ↵Gravatar A. Unique TensorFlower2017-03-23
| | | | | | parse_single_example. Change: 151038761
* Add comments for Android NDK settings in WORKSPACE.Gravatar Andrew Harp2017-03-23
| | | | Change: 151036764
* Fix typo referring to provided strategy function byte_size_load_fn.Gravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151036671
* Adds continuous_train_and_evaluate in Experiment.Gravatar Jianwei Xie2017-03-23
| | | | Change: 151031162
* Support int64 to float mapping in MutableHashTableGravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151031147
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151018201
* Remove inflow mention from feature_column_ops docstringsGravatar David Soergel2017-03-23
| | | | Change: 151017665
* tf.string_to_number: Add support for int64 and float64.Gravatar A. Unique TensorFlower2017-03-23
| | | | Change: 151015024
* Fix misleading text in TensorBoard tutorial.Gravatar Brennan Saeta2017-03-23
| | | | Change: 151012934
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-09 04:25:35 +0000