aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
...
* GPU registration for resource scatter addGravatar Alexandre Passos2017-07-17
| | | | PiperOrigin-RevId: 162289810
* Add support for cublas<t>geam to cuda_solvers kernel helpers.Gravatar Eugene Brevdo2017-07-17
| | | | PiperOrigin-RevId: 162284275
* TFTS: LSTM model-building exampleGravatar Allen Lavoie2017-07-17
| | | | PiperOrigin-RevId: 162282906
* Adds multi-class (Crammer-Singer) hinge loss to tf.contrib.kernel_methodsGravatar Petros Mol2017-07-17
| | | | PiperOrigin-RevId: 162277054
* Restore previous IrEmitter::HandleSlice() behavior.Gravatar A. Unique TensorFlower2017-07-17
| | | | | | | | | END_PUBLIC BEGIN_PUBLIC Automated g4 rollback of changelist 162249449 PiperOrigin-RevId: 162276286
* [XLA] Allow logical buffer coloring to use liveness information.Gravatar A. Unique TensorFlower2017-07-17
| | | | PiperOrigin-RevId: 162275084
* TensorForest v4 kernel tests for proto-based model. Also some stuff to ↵Gravatar A. Unique TensorFlower2017-07-17
| | | | | | include decision_trees in pip package properly. PiperOrigin-RevId: 162273792
* Avoid the duplication in debug_options_flags.cc by generalizing ↵Gravatar A. Unique TensorFlower2017-07-17
| | | | | | tensorflow::Flag. PiperOrigin-RevId: 162271241
* CPU IR Emitter: Lower slice operations with memcpy.Gravatar A. Unique TensorFlower2017-07-17
| | | | | | | | | | | | | | | This changes the way the CPU backend emits slice operations so that the maximal contiguous blocks get copied with memcpy. Without this change, LLVM's optimizers prefer to unroll the elementwise loops emitted for doing slice. This leads to increased code size. With this change, the inliner only inlines the memcpy when it is small enough. This change also adds a test that exercises the logic to find maximal contiguous blocks. PiperOrigin-RevId: 162249449
* Add a gRPC client for profiling TPU (contrib/tpu/profiler/)Gravatar Eric Liu2017-07-17
| | | | | | | This contains a gRPC client that starts/stops tracing and processes/stores the result trace data into a TensorBoard log directory. This also exposes trace_events proto classes via tf.contrib.tpu.profiler public API so that TensorBoard's profile plugin can process and visualize the profile. PiperOrigin-RevId: 162247333
* Fix style issue in tfdbg documentation.Gravatar A. Unique TensorFlower2017-07-17
| | | | PiperOrigin-RevId: 162244864
* Create ScopedTFStatus to manage TF_Status lifetime.Gravatar Skye Wanderman-Milne2017-07-17
| | | | | | | | | In addition, this change: - Adds a SWIG typemap to automatically unwrap ScopedTFStatus's for TF_Status* inputs - Creates a new c_api_util.py file - Switches a few TF_Status users to use raise_exception_on_not_ok_status() instead PiperOrigin-RevId: 162243205
* Make TensorForest V4 the default implementation.Gravatar A. Unique TensorFlower2017-07-17
| | | | PiperOrigin-RevId: 162230001
* Minor cleanupGravatar A. Unique TensorFlower2017-07-17
| | | | PiperOrigin-RevId: 162222071
* Split out new while_loop_test.cc from c_api_test.ccGravatar Skye Wanderman-Milne2017-07-17
| | | | | | | | This change also separates shared functionality into c_test_util.h/cc. This brings c_api_test.cc to a mere 1715 LOC (further splits can be more easily done now too). PiperOrigin-RevId: 162216399
* Move cluster initialization to happen during training using training hooks ↵Gravatar A. Unique TensorFlower2017-07-17
| | | | | | instead of depending on variable initialization logic. PiperOrigin-RevId: 162215450
* Adds evaluation_hooks in EstimatorSpec.Gravatar Jianwei Xie2017-07-15
| | | | PiperOrigin-RevId: 162078550
* FileBlockCache is now a filesystem-wide shared cache with a configurable ↵Gravatar A. Unique TensorFlower2017-07-14
| | | | | | maximum size (in total bytes of cached block contents). The default maximum cache size is equal to the block size, and can be overridden by the environment variable GCS_READ_CACHE_MAX_SIZE_MB. In addition, deleting files now flushes them from the block cache, and renaming files flushes both the source and destination files from the block cache. PiperOrigin-RevId: 162041685
* Add missing grpcio install to install_python3.5_pip_packages.shGravatar Shanqing Cai2017-07-14
| | | | PiperOrigin-RevId: 162040682
* Change Affine bijector to use LinearOperator instead of OperatorPD.Gravatar A. Unique TensorFlower2017-07-14
| | | | PiperOrigin-RevId: 162034086
* Update Keras API exports.Gravatar Francois Chollet2017-07-14
| | | | PiperOrigin-RevId: 162030530
* Add unit-test for questions:Gravatar Joshua V. Dillon2017-07-14
| | | | | | | - http://stackoverflow.com/q/45109305 - #10766 PiperOrigin-RevId: 162026912
* In colocation op device checking, canonicalize the names soGravatar Vijay Vasudevan2017-07-14
| | | | | | | | | | | | | /device:GPU:0 compares to /gpu:0 as one would expect. Adds a test to ensure that the warning message isn't printed by checking the resulting device name. No behavior change is expected here; all this does is change to share the canonical name and remove the warning. Fixes #11484. PiperOrigin-RevId: 162024152
* Fix race in record_yielder.Gravatar A. Unique TensorFlower2017-07-14
| | | | PiperOrigin-RevId: 162022981
* API and an example of two-level quantile accumulator.Gravatar A. Unique TensorFlower2017-07-14
| | | | PiperOrigin-RevId: 162022193
* Add overlap_and_add to tf.contrib.signal.Gravatar RJ Ryan2017-07-14
| | | | | | This is the inverse operation of tf.contrib.signal.frame. This is implemented using GPU-capable ops and supports gradients. PiperOrigin-RevId: 162017464
* Fix vector/scalar typos in sequence example parser error messages. Old code ↵Gravatar A. Unique TensorFlower2017-07-14
| | | | | | erroneously claimed that scalar arguments should be vectors. PiperOrigin-RevId: 162009915
* Add GPU kernel registrations for int64 tf.concat, tf.stackGravatar A. Unique TensorFlower2017-07-14
| | | | PiperOrigin-RevId: 162008248
* Add no_oss and nomac filtering to run_pip_test.shGravatar Shanqing Cai2017-07-14
| | | | PiperOrigin-RevId: 162002161
* Add new features to tf.contrib.signal.frames:Gravatar RJ Ryan2017-07-14
| | | | | | | | | | | - Support arbitrary rank tensors. - Add an `axis` parameter for framing any axis. - Add a `pad_end` and `pad_value` parameter for controlling the signal padding. Padding is disabled by default to avoid a copy of a potentially large Tensor. - Support shape inference of resulting framed tensor. - Expand the tests, including tests for the gradients. - Since `frames` is a noun and not a verb, rename `frames` to `frame`. PiperOrigin-RevId: 161998921
* Add stats collection for ops used within functions.Gravatar A. Unique TensorFlower2017-07-14
| | | | PiperOrigin-RevId: 161997142
* slice_test: Add a few tests for strided slices andGravatar A. Unique TensorFlower2017-07-14
| | | | | | | | | | consolidate R1 tests. * Fix SliceUnaligned1024In4096Values, which was using unitialized data. * Also fix SliceR2Test to respect the layout parameter. PiperOrigin-RevId: 161991313
* Disable contrib/keras:data_utils_test on ASAN since it sometimes times out.Gravatar A. Unique TensorFlower2017-07-14
| | | | PiperOrigin-RevId: 161990256
* Better gating of TensorForest pruning code.Gravatar A. Unique TensorFlower2017-07-14
| | | | PiperOrigin-RevId: 161981809
* Remove TODO comment.Gravatar A. Unique TensorFlower2017-07-14
| | | | PiperOrigin-RevId: 161979883
* [XLA] Make computation_layouts_ protected.Gravatar Blake Hechtman2017-07-14
| | | | PiperOrigin-RevId: 161972749
* Adds shards in head_test to prevent test timeout.Gravatar A. Unique TensorFlower2017-07-14
| | | | PiperOrigin-RevId: 161967860
* Document tf.Learn SavedModel export.Gravatar David Soergel2017-07-14
| | | | PiperOrigin-RevId: 161967071
* Switch TensorForestEstimator to tf.learn heads lib.Gravatar A. Unique TensorFlower2017-07-14
| | | | PiperOrigin-RevId: 161949540
* Slightly change the test case that tests wrap-around for DynamicSlice with R3.Gravatar A. Unique TensorFlower2017-07-14
| | | | | | It still tests wrap-around behavior, but in the most minor dimension instead of in the second most minor dimension. PiperOrigin-RevId: 161924370
* Add a first draft of the "graphs and sessions" section of the programmers' ↵Gravatar Derek Murray2017-07-13
| | | | | | guide. PiperOrigin-RevId: 161917422
* Stop running some tests on ASAN that are timing out.Gravatar A. Unique TensorFlower2017-07-13
| | | | PiperOrigin-RevId: 161906344
* Refactoring the TPUEstimator.Gravatar Jianwei Xie2017-07-13
| | | | PiperOrigin-RevId: 161905196
* No reason to run build_docs_test through the sanitizers.Gravatar A. Unique TensorFlower2017-07-13
| | | | PiperOrigin-RevId: 161903640
* Adds a file cache to the GCS filesystem, with configurable max staleness for ↵Gravatar A. Unique TensorFlower2017-07-13
| | | | | | file contents. Maximum staleness is specified in seconds in the GCS_READ_CACHE_MAX_STALENESS environment variable, and defaults to 0 (indicating that staleness is not tolerated in a newly opened file). Staleness is measured from the arrival time of the first block in the file's block cache. PiperOrigin-RevId: 161897769
* Go: Update generated wrapper functions for TensorFlow ops.Gravatar A. Unique TensorFlower2017-07-13
| | | | PiperOrigin-RevId: 161897268
* Update ops-related pbtxt files.Gravatar A. Unique TensorFlower2017-07-13
| | | | PiperOrigin-RevId: 161896495
* Fix "Converting to Python 3" error for check_load_py_test.Gravatar A. Unique TensorFlower2017-07-13
| | | | PiperOrigin-RevId: 161896415
* [XLA] Pass shape/layout information in calls to the CPU runtime routines.Gravatar A. Unique TensorFlower2017-07-13
| | | | | | | | | | | Previously the CPU runtime wouldn't know how the data that was being outfed was laid out by the XLA LayoutAssignment pass, which could result in transposed-value results. This also allows us to validate the contract between the host program and the compiled XLA program with (reified) runtime type checks. PiperOrigin-RevId: 161895093
* Adds gzip support to TextLineDataset.Gravatar A. Unique TensorFlower2017-07-13
| | | | | | - TextLineDataset now accepts the compression_type argument. PiperOrigin-RevId: 161892408
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-05 08:58:11 +0000