1. fc7651f Avoid using SSE_OR_PORTABLE am: 9e3cbbdd7c by Miao Wang · 3 weeks ago master
  2. 9e3cbbd Avoid using SSE_OR_PORTABLE by Miao Wang · 3 weeks ago
  3. 8fd0377 Merge remote-tracking branch 'aosp/upstream-master' into tfl_rebase by Miao Wang · 3 weeks ago
  4. e6330c9 Merge remote-tracking branch 'aosp/upstream-master' into tfl_rebase by Miao Wang · 5 weeks ago
  5. 1149df1 Support quantization for tfl.strided_slice by Feng Liu · 5 weeks ago
  6. 716c07a Make sure the buffers in NNAPI delegate shared memory pool are 16 bytes aligned. by Miao Wang · 5 weeks ago
  7. 40a34d1 Converts hardswish subgraphs into atomic ops. by A. Unique TensorFlower · 5 weeks ago
  8. fdeee73 Update GraphDef version to 124. by A. Unique TensorFlower · 5 weeks ago
  9. c7795c8 compat: Update forward compatibility horizon to 2019-08-11 by A. Unique TensorFlower · 5 weeks ago
  10. 6e7facf Remove SplitAndParseAsFloats from tensorflow::strings API. by Gunhan Gulsoy · 5 weeks ago
  11. 24c5f47 Fuse Add into the proceding FullyConnected when there is a bias by Feng Liu · 5 weeks ago
  12. 119303c Make sure the buffers in NNAPI delegate shared memory pool are 16 bytes aligned. by Miao Wang · 5 months ago
  13. 516f7aa Fully qualify tstring in TF_CALL_tstring macros. by Dero Gharibian · 5 weeks ago
  14. 029fb28 Add patterns to switch BiasAdd/AddV2 and Mul by Feng Liu · 5 weeks ago
  15. 61f7831 Refactor DenseElementAttr::getValues methods to return full ranges for splats. by River Riddle · 5 weeks ago
  16. 0f64a9c Cleanup TF executor dialect island coarsening pass (NFC). by Andy Ly · 5 weeks ago
  17. 9aad37c Migrated tensorflow/core/platform/ to use tstring. by Dero Gharibian · 5 weeks ago
  18. c63c04a Rename namespace TFExecutor to tf_executor in IslandCoarsening pass (NFC) by Mehdi Amini · 5 weeks ago
  19. 5c06ca4 Remove unused header file by Ashwin Murthy · 5 weeks ago
  20. 2e7fa3d Correctly convert const int8 weights to uint8 for NNAPI. by A. Unique TensorFlower · 5 weeks ago
  21. 0a9f6d0 Remove unnecessary const & for Type and also remove unnecessary 'template'. by Renjie Liu · 5 weeks ago
  22. af8d4fd Ensure we can propagate error properly when setting "TF_ENABLE_EAGER_CLIENT_STREAMING_ENQUEUE=false". by Xiao Yu · 5 weeks ago
  23. b4bf76a Migrated tensorflow/core/framework/ to use tstring. by Dero Gharibian · 5 weeks ago
  24. 5e307a0 Move annotation's thread_local storage to .cc file and clean up dependencies to avoid a duplicate symbol error. by A. Unique TensorFlower · 5 weeks ago
  25. 90169fe [XLA] Turn gathers of very small vectors into many selects. by Blake Hechtman · 5 weeks ago
  26. ccd578f Update GraphDef version to 123. by A. Unique TensorFlower · 5 weeks ago
  27. d37d29b compat: Update forward compatibility horizon to 2019-08-10 by A. Unique TensorFlower · 5 weeks ago
  28. b3016c4 Merge pull request #31001 from ROCmSoftwarePlatform:google-upstream-pr-rnn-ops by TensorFlower Gardener · 5 weeks ago
  29. 08ee387 Remove redundant const reference on llvm::ArrayRef by Ashwin Murthy · 5 weeks ago
  30. ec2d7ca Fix TF select op code path. by Tian Lin · 5 weeks ago
  31. 8190fa8 NFC: Standardize the terminology used for parent ops/regions/etc. by River Riddle · 5 weeks ago
  32. cb01a29 Fix tpu_ops.all_to_all op output shape. by A. Unique TensorFlower · 5 weeks ago
  33. e6dc56c TFLite GPU OpenGL: Add specialization of SOFTMAX 1x1. by Juhyun Lee · 5 weeks ago
  34. 9e88516 NFC: Refactoring PatternSymbolResolver into SymbolInfoMap by Lei Zhang · 5 weeks ago
  35. bf62fce Automated rollback of commit 23e33f871b2bf2879b40ebf3b883e104f30f389b. Revert #31450. by Anna R · 5 weeks ago
  36. 019da8c Turn streaming rpc off by default. When stream rpc is off, fall back to use non-streaming rpc. by Xiao Yu · 5 weeks ago
  37. f4e6f6b Update ops-related pbtxt files. by A. Unique TensorFlower · 5 weeks ago
  38. 1c58153 [TFLite] Add a ModulePass to prune unexported functions before legalization to TFLite dialect. by Ashwin Murthy · 5 weeks ago
  39. fa7262e Tensor tracer: Adding full-tensor summary writing capability to tensor tracer. by A. Unique TensorFlower · 5 weeks ago
  40. 4faa97c Strip numerical ports from TPU paths starting with /bns. by Revan Sopher · 5 weeks ago
  41. 1590fbf NFC: Update usages of OwningRewritePatternList to pass by & instead of &&. by River Riddle · 5 weeks ago
  42. c271613 Added medium post and github links to posenet website. by Eileen Mao · 5 weeks ago
  43. d86a92d [Tensorflow Metrics] Add metrics to track whether tf2 behavior is turned on by environmental flags. by A. Unique TensorFlower · 5 weeks ago
  44. b69d643 TFLite GPU Delegate: prepare for registering custom ops in Metal. by A. Unique TensorFlower · 5 weeks ago
  45. 5228e5f Removing unused function AdjustFilenameForLogging. by Brian Zhao · 5 weeks ago
  46. 1960c9d Prepare convert_to_constants.py for tensor equality changes. by Saurabh Saxena · 5 weeks ago
  47. 9e3794f A ragged to dense op for directly calculating tensors. by A. Unique TensorFlower · 5 weeks ago
  48. 35e250b Add model metadata for profiling. by A. Unique TensorFlower · 5 weeks ago
  49. d221f8e Not apply new eq change to graph function building mode by Yanhua Sun · 5 weeks ago
  50. c52f26a [SE] Change LoadModuleFromCuBin and LoadPtx to return Status instead of bool by George Karpenkov · 5 weeks ago
  51. 034be38 Automated rollback of commit c9552455ab76ec1b4c15bcac5aa53b85b7463175 by Ayush Dubey · 5 weeks ago
  52. 331c10a Merge pull request #31469 from Intel-tensorflow:guizili/fix_klockworks by TensorFlower Gardener · 5 weeks ago
  53. 42bab29 NFC: Clean up XLA op definitions by A. Unique TensorFlower · 5 weeks ago
  54. 5293302 Update Eigen to https://bitbucket.org/eigen/eigen/commits/8071cda5714d7f454205b60b2f851986d53e8f33 by A. Unique TensorFlower · 5 weeks ago
  55. e545a51 Switch back to use easy_install to install pip2 and pip3, but let pip be upgraded to the latest by not specifying the version. by Yifei Feng · 5 weeks ago
  56. b78d23c Update the TensorInfo protobuf message with an encoding for composite tensors; and update SavedModel to use this new encoding. by Edward Loper · 5 weeks ago
  57. 23e33f8 Merge pull request #31450 from trevor-m:tmorris_flatten_int64 by TensorFlower Gardener · 5 weeks ago
  58. dac283f Extend the subclass model testing utils to support ragged/sparse input. by A. Unique TensorFlower · 5 weeks ago
  59. 39e9aa1 In UpdateTFE_ContextWithServerDef(), call GrpcServer::Start() after all other initialization. by Derek Murray · 5 weeks ago
  60. 513e06d export enable_tensor_equality disable_tensor_equality so that users have way to opt-in and opt-out explicitly by Yanhua Sun · 5 weeks ago
  61. ae4267e Disable gan tests as they are failing on contrib continuous. by Gunhan Gulsoy · 5 weeks ago
  62. 8c4252c Add specific error for functions capturing Keras learning phase, and fix keras saving tests. by Katherine Wu · 5 weeks ago
  63. f5ab4f8 Fix compilation error on arm32 by Benoit Jacob · 5 weeks ago
  64. 1cc1461 Update training v2 execution function to return batch size. by Scott Zhu · 5 weeks ago
  65. 25310d0 Add legalization pattern for tf.SelectV2 op. by Haoliang Zhang · 5 weeks ago
  66. c955245 Automated rollback of commit 11a5d679a4da5cda405a66f267368ee45fa980c7 by Ayush Dubey · 5 weeks ago
  67. fafaf6b Ruy: Correct an include. PiperOrigin-RevId: 262613092 by Alex Stark · 5 weeks ago
  68. 6e3a6bb Disable two bfloat16 tests on Windows. by Reed Wanderman-Milne · 5 weeks ago
  69. aed9360 Automated rollback of commit 8412e4920296dd3df0ba1a99e4f3f783f74fcda2 by A. Unique TensorFlower · 5 weeks ago
  70. d24f662 Implement __repr__ for LossScale subclasses. by Reed Wanderman-Milne · 5 weeks ago
  71. 26220ec Improve Layer docstrings in regards to autocasting. by Reed Wanderman-Milne · 5 weeks ago
  72. 0aca78d Rename TFLite metal_delegate functions. by Yu-Cheng Ling · 5 weeks ago
  73. 3f5348e Check "cond" and "body" are defined in tf.While verifier. by Jing Pu · 5 weeks ago
  74. 3eba235 Fold multiplier even the operands are not broadcast-compatible by Feng Liu · 5 weeks ago
  75. 369a886 Add the `strides` param to the slice operation semantics doc by Sanjoy Das · 5 weeks ago
  76. d936bf76 Translation to LLVM IR: use LogicalResult instead of bool by A. Unique TensorFlower · 5 weeks ago
  77. c02d99f [XLA] [AlgebraicSimplifier] Reuse visitor across computations. by Yunxing Dai · 5 weeks ago
  78. 7f70878 Adds support for FusedBatchNormV3 in converter by Sachin Joglekar · 5 weeks ago
  79. 0cdf225 Only enable graph rewrite for RNN layer in v2 mode (outmost eager context). by Scott Zhu · 5 weeks ago
  80. 0c38681 Fix the test for std::aligned_alloc availability. This enables compilation with C++17 on Apple by A. Unique TensorFlower · 5 weeks ago
  81. 8b639d3 Extracts NNAPIDelegateKernel from nnapi_delegate.cc by Stefano Galarraga · 5 weeks ago
  82. 1cd4820 LLVM dialect and translation: support global strings by A. Unique TensorFlower · 5 weeks ago
  83. 5b4fe54 In the underflow case of QuantizeMultiplier, when an exponent smaller than -31 would be produced, which would mean that the corresponding right-shifts would always produce zero anyway but would require additional code to handle by Benoit Jacob · 5 weeks ago
  84. c94c683 Rewrite the handling of threads==1, so it's a little more readable, and gets compiled with -O3 in a way that puts this case at the start of the function instead of at the end, which for a mysterious reason results in more stable performance. by Benoit Jacob · 5 weeks ago
  85. 977f213 Translation to LLVM: support llvm.global by A. Unique TensorFlower · 5 weeks ago
  86. 5a174fe Avoid needlessly aborting when this graph transformation should just bail and allow continuing. by Benoit Jacob · 5 weeks ago
  87. 6c526e0 Specify -O3 and, on ARM32, -mfpu=neon as rule copts, for all our binary rules. by Benoit Jacob · 5 weeks ago
  88. a79c52e Added default move constructor and assignment operator for tstring. by Dero Gharibian · 5 weeks ago
  89. ede03c7 Merge pull request #31466 from richardbrks:publicJNI by TensorFlower Gardener · 5 weeks ago
  90. 3737675 Disable AVX512 on __APPLE__ for now to unbreak the build. by Benoit Jacob · 5 weeks ago
  91. 8bb9d50 External library name mangling support for linalg. by Nicolas Vasilache · 5 weeks ago
  92. a475e3e Allow linalg.view to change the underlying elemental type. by Nicolas Vasilache · 5 weeks ago
  93. de9771e Add a higher-order vector.outerproduct operation in MLIR by Nicolas Vasilache · 5 weeks ago
  94. c22f09d Merge pull request #31405 from saket404:typo_fix by TensorFlower Gardener · 5 weeks ago
  95. eecba4e Merge pull request #30660 from ROCmSoftwarePlatform:google_upstream_xla_amdgpu_frontend by TensorFlower Gardener · 5 weeks ago
  96. 2d8d173 Add a higher-order vector.extractelement operation in MLIR by Nicolas Vasilache · 5 weeks ago
  97. b07072c Merge pull request #27367 from ANSHUMAN87:add-test-warn-fix by TensorFlower Gardener · 5 weeks ago
  98. ba9c03a Add support for vector ops in the LLVM dialect by Nicolas Vasilache · 5 weeks ago
  99. 964bbd4 LLVM Dialect: introduce llvm.global by A. Unique TensorFlower · 5 weeks ago
  100. ed0eee6 BlockLSTM and LSTMBlockCell are now generic wrt to the gate layout by Sergei Lebedev · 5 weeks ago