aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* [Go]: Support device annotations when constructing graphs.Gravatar Asim Shankar2018-07-11
| | | | PiperOrigin-RevId: 204225504
* Enrich update ops from inputsGravatar Zhenyu Tan2018-07-11
| | | | PiperOrigin-RevId: 204223077
* [tflite] Add macro registration for type string to typeToTfLiteType.Gravatar Eugene Brevdo2018-07-11
| | | | PiperOrigin-RevId: 204210338
* Add support for microcontroller-scale audio speech modelsGravatar Pete Warden2018-07-11
| | | | PiperOrigin-RevId: 204204001
* Add a few naming guidelines for the converter library.Gravatar Dan Moldovan2018-07-11
| | | | PiperOrigin-RevId: 204199604
* Merge pull request #20685 from yongtang:06252018-roll-fast-bound-checkGravatar TensorFlower Gardener2018-07-11
|\ | | | | | | PiperOrigin-RevId: 204199229
* | Small optimization on NeonMatrixBatchVectorMultiplyAccumulate and ↵Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | NeonVectorBatchVectorCwiseProductAccumulate. PiperOrigin-RevId: 204196998
* | Merge pull request #20705 from case540:cherry_pick_find_cmd_fixGravatar TensorFlower Gardener2018-07-11
|\ \ | | | | | | | | | PiperOrigin-RevId: 204196525
* | | Exposes device_assignment in model_fn and implements full_replicate infeed mode.Gravatar Youlong Cheng2018-07-11
| | | | | | | | | | | | PiperOrigin-RevId: 204194591
* | | Add full cluster resolver package to contrib_py rather than the ↵Gravatar Frank Chen2018-07-11
| | | | | | | | | | | | | | | | | | implementation-less skeleton. PiperOrigin-RevId: 204188173
* | | Merge pull request #20617 from aaroey:fix_unused_statusGravatar TensorFlower Gardener2018-07-11
|\ \ \ | | | | | | | | | | | | PiperOrigin-RevId: 204186827
* | | | Expose the bfloat16 in tf.contrib.tpu.Gravatar Jianwei Xie2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204183931
* | | | [XLA] Enable tests to force result layouts for tuple results.Gravatar Michael Kuperstein2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204180483
* | | | Internal change.Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204177704
* | | | Add documentation to TF Lite demo example appGravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204176713
* | | | Always append ExamplesPerSecond hook in TPUEstimatorGravatar Yanan Cao2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204164915
* | | | Do not mark constants as guaranteed constants.Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204162855
* | | | [tf.data] Remove outdated comment.Gravatar Jiri Simsa2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204160548
* | | | Implements tf.metrics.f1_score as the maximum f1 score across different ↵Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | thresholds. PiperOrigin-RevId: 204159841
* | | | Merge pull request #20641 from taehoonlee:fix_typosGravatar TensorFlower Gardener2018-07-11
|\ \ \ \ | | | | | | | | | | | | | | | PiperOrigin-RevId: 204158338
* | | | | Allow LinearOperatorLowerTriangular and LinearOperatorLowRankUpdate to ↵Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | accept complex types. PiperOrigin-RevId: 204156712
* | | | | Forward -w down to nvcc.Gravatar Justin Lebar2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This makes --copt=-w silence warnings both in the host compiler and in the CUDA compiler. PiperOrigin-RevId: 204156508
* | | | | Add average prediction, average label to regression head.Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204154837
| | | * | Match for path instead of nameGravatar Yifei Feng2018-07-11
| |_|/ / |/| | |
* | | | Split out lib/png into a separate library from the rest of TensorFlow to avoidGravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | unnecessary transitive dependencies. PiperOrigin-RevId: 204154526
* | | | Rename LLVMIRGenTestBase to LlvmIrGenTestBase as per Google's C++ style; NFCGravatar Sanjoy Das2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204153252
* | | | Improve error handling and usability of Python interpreterGravatar Andrew Selle2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Break dependency on tensorflow platform and logging and absl - Propagate exceptions that capture the TensorFlow lite errors in a buffer. PiperOrigin-RevId: 204148724
* | | | Expose quant_delay arg in experimental_create_eval_graph function.Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204148516
* | | | [TF:XLA] Add implementation of ResourceApplyCenteredRMSProp.Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | Add support to XlaOpKernelContext for accessing inputs by name . PiperOrigin-RevId: 204148428
* | | | [tf.data] Fixing a subtle use-after-free bug in the optimization logic.Gravatar Jiri Simsa2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204146100
* | | | Minor doc updates.Gravatar Dan Moldovan2018-07-11
| | | | | | | | | | | | | | | | PiperOrigin-RevId: 204144214
* | | | Alias non-entry parameters with temp buffersGravatar Sanjoy Das2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Parameters to while nodes, for instance, do alias with temps. I haven't directly encountered this bug on HEAD, but a proper fix for b/111116907 exposes this. PiperOrigin-RevId: 204143492
* | | | Merge pull request #20654 from alexbw:patch-3Gravatar TensorFlower Gardener2018-07-11
|\ \ \ \ | | | | | | | | | | | | | | | PiperOrigin-RevId: 204142076
* | | | | [tf.data / Bigtable] Use builder pattern for client optionsGravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204140674
* | | | | Merge pull request #20400 from yongtang:06202018-nasmGravatar TensorFlower Gardener2018-07-11
|\ \ \ \ \ | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204139191
* | | | | | [XLA:GPU] s/llvm_ir::IrArray/IrArray/ in ir_emitter_unnested.Gravatar Justin Lebar2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Less visual noise. PiperOrigin-RevId: 204139183
* | | | | | Add user-defined initializers to the IndyLSTMCell.Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204137901
* | | | | | [XLA:GPU] Cleanups to fused 021 transpose implementation.Gravatar Justin Lebar2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Fix typos. - Clarify comments. - Reduce nesting in a few places. - Add asserts that this code is dealing with specifically a loop fusion. - Rename some functions. In particular, it's confusing to have a function with a generic name like EmitCodeWithBoundCheck that actually is specialized to a tiled implementation. - Remove statement expression (GCC language extension), replacing it with an IIFE. - Don't refer to shared-memory tile space as "buffer" without other qualifying words, since that's ambiguous with what XLA refers to as a "buffer". - Use llvm::cast instead of static_cast. - Comply with style guide naming rules for compile-time constants (kFoo). - Use c_accumulate instead of std::accumulate. - Put std::function parameter at the end of the param list. This lets us cleanly embed the lambda into the call because of how clang-format formats such calls. (I think this one is possibly the most helpful change in this patch, as it suddenly makes clear to me the way that we use two calls to emit_tiled_elemental_code_with_bounds_check to emit the code.) PiperOrigin-RevId: 204134102
* | | | | | Improvements to testing of proto decode / encode ops.Gravatar Jiri Simsa2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204132868
* | | | | | Upgrade docker Bazel version to 0.14.1Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204131441
* | | | | | [tf.data] Add support for handling function library definition optimizations.Gravatar Jiri Simsa2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204128363
* | | | | | Skeleton for integration tests.Gravatar Dan Moldovan2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204114884
* | | | | | [Java]: Release 1.9.0Gravatar Asim Shankar2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204114708
* | | | | | Fix DCHECK off-by-one error.Gravatar Adrian Kuegel2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We should use DCHECK_GE instead of DCHECK_GT. PiperOrigin-RevId: 204107900
* | | | | | Add CROSSTOOL for Windows GPU build with nvcc compilerGravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After this change, to build TensorFlow with GPU support on Windows, you just need to follow the same procedure as Linux. Also re-enable remote cache since the bug in b/111106393 has been fixed by the new wrapper script. The clean-up of the wrapper scripts also made the build around 5 mins faster than using Bazel's msvc wrapper script. PiperOrigin-RevId: 204105368
* | | | | | Update downloadable clang to r336424Gravatar Ilya Biryukov2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204105366
* | | | | | Clamp gather indices instead of wrap-around.Gravatar Adrian Kuegel2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | So far, the behavior for out of bounds cases is undefined as long as it does not crash. Clamping is cheaper because we don't need a URem operation, and for DynamicSlice we already do that. PiperOrigin-RevId: 204096995
* | | | | | Avoid breakages of the TF weighted quantiles library if used with values ↵Gravatar A. Unique TensorFlower2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | being non-numeric types such as strings. PiperOrigin-RevId: 204087922
* | | | | | Disable XlaSort and TopKV2 ops for the XLA CPU/GPU backends.Gravatar Adrian Kuegel2018-07-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | TopKV2 is implemented by using XlaSort, and XlaSort is not implemented yet on CPU and GPU. PiperOrigin-RevId: 204082873
* | | | | | Update estimators section for getting started.Gravatar Billy Lamberta2018-07-10
| | | | | | | | | | | | | | | | | | | | | | | | PiperOrigin-RevId: 204071680
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-29 10:22:22 +0000