Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Mar 4, 2020

• Avoid baking random seeds into RandomDataset and SamplingDataset.

  Show description for 97f595f
  

  aaudiber

  authored and

  tensorflower-gardener

  committed
  97f595f

  Copy full SHA for 97f595f

• Add test for recovering from compile errors w/ TPUStrategy.

  Show description for f6793de
  

  kenfranko

  authored and

  tensorflower-gardener

  committed
  f6793de

  Copy full SHA for f6793de

• Eliminate tf.where call in updating running mean and average for batch norm in Keras. This unlocks the improvements promised by fusing updates into the CPU and GPU kernels.

  Show description for f7d4c7f
  tensorflower-gardener

  committed
  f7d4c7f

  Copy full SHA for f7d4c7f

• Kokoro related change for switching `nighlty-2.x` to the `nightly` and removing older `nightly` which was pointing to nightly-1.15.

  Show description for 8f39997
  tensorflower-gardener

  committed
  8f39997

  Copy full SHA for 8f39997

• [XLA:Python] Guard NCCL usage by if_cuda rather than if_nccl.

  Show description for ca8d8fb
  

  hawkinsp

  authored and

  tensorflower-gardener

  committed
  ca8d8fb

  Copy full SHA for ca8d8fb

• Pull partition (row_splits, value_rowid, et cetera) of a RaggedTensor into a separate object. This allows one RowPartition to be shared by multiple RaggedTensor objects.

  Show description for 6c26c99
  tensorflower-gardener

  committed
  6c26c99

  Copy full SHA for 6c26c99

• Disable flaky test.

  Show description for 0b0432a
  tensorflower-gardener

  committed
  0b0432a

  Copy full SHA for 0b0432a

• Refactoring of more existing snapshot code into common helper library

  Show description for 163055b
  

  Frank Chen

  authored and

  tensorflower-gardener

  committed
  163055b

  Copy full SHA for 163055b

• Add tests for tf.one_hot() ragged tensor handling.

  Show description for 90d563c
  

  edloper

  authored and

  tensorflower-gardener

  committed
  90d563c

  Copy full SHA for 90d563c

• Switch tf2tensorrt target to use pybind11

  Show description for 2e7bd97
  

  reedwm

  authored and

  tensorflower-gardener

  committed
  2e7bd97

  Copy full SHA for 2e7bd97

• [MLIR][XLA] Lower xla_hlo.sqrt from HLO -> LHLO -> Linalg.

  Show description for 2b6eab2
  

  pifon2a

  authored and

  tensorflower-gardener

  committed
  2b6eab2

  Copy full SHA for 2b6eab2

• [ModelBuilder] Use properly padded Vector in MLIR's CRunnerUtils.h

  Show description for 1efba0d
  

  nicolasvasilache

  authored and

  tensorflower-gardener

  committed
  1efba0d

  Copy full SHA for 1efba0d

• Add Lit test configurations to support AOT compilation.

  Show description for 5b3d404
  

  lucarfox

  authored and

  tensorflower-gardener

  committed
  5b3d404

  Copy full SHA for 5b3d404

• Update tflite writer lib for recently added half-pixel-centers flag to resize-bilinear op.

  Show description for 70069b0
  tensorflower-gardener

  committed
  70069b0

  Copy full SHA for 70069b0

• Fix bugs in tf.ragged.cross that impacts windows:

  Show description for f6ff46c
  

  edloper

  authored and

  tensorflower-gardener

  committed
  f6ff46c

  Copy full SHA for f6ff46c

• Run tf.ragged.cross tests in graph & eager mode.

  Show description for c060d9d
  

  edloper

  authored and

  tensorflower-gardener

  committed
  c060d9d

  Copy full SHA for c060d9d

• Internal change for TFRT.

  Show description for a709d28
  tensorflower-gardener

  committed
  a709d28

  Copy full SHA for a709d28

• [XLA][MLIR] Reduce code bloat for LHLO->STD and HLO->STD patterns.

  Show description for 4aaabc8
  

  pifon2a

  authored and

  tensorflower-gardener

  committed
  4aaabc8

  Copy full SHA for 4aaabc8

• Go: Update generated wrapper functions for TensorFlow ops.

  Show description for 8a53e35
  tensorflower-gardener

  committed
  8a53e35

  Copy full SHA for 8a53e35

• Properly print the `msg` attribute when assertNotEqual tests fail.

  Show description for 34197e1
  

  ZacharyGarrett

  authored and

  tensorflower-gardener

  committed
  34197e1

  Copy full SHA for 34197e1

• Go: Update generated wrapper functions for TensorFlow ops.

  Show description for b747c91
  tensorflower-gardener

  committed
  b747c91

  Copy full SHA for b747c91

• Add internal visibility.

  Show description for 9518f70
  

  chsigg

  authored and

  tensorflower-gardener

  committed
  9518f70

  Copy full SHA for 9518f70

• Lower log op from HLO all the way to Linalg

  Show description for 2ef1349
  

  akuegel

  authored and

  tensorflower-gardener

  committed
  2ef1349

  Copy full SHA for 2ef1349

• Add tests to check the initialization of the input tensor.

  Show description for 5731c94
  

  dayeongl

  authored and

  tensorflower-gardener

  committed
  5731c94

  Copy full SHA for 5731c94

• compat: Update forward compatibility horizon to 2020-03-04

  Show description for 8f080db
  tensorflower-gardener

  committed
  8f080db

  Copy full SHA for 8f080db

• Merge pull request #36699 from mitchellvitez:input_cast_warning_typos

  Show description for 02f663e
  tensorflower-gardener

  committed
  02f663e

  Copy full SHA for 02f663e

• Go: Update generated wrapper functions for TensorFlow ops.

  Show description for a1ead75
  tensorflower-gardener

  committed
  a1ead75

  Copy full SHA for a1ead75

• Clean up profiler when exception happens.

  Show description for 58f761f
  tensorflower-gardener

  committed
  58f761f

  Copy full SHA for 58f761f

• Internal change

  Show description for 77842e6
  tensorflower-gardener

  committed
  77842e6

  Copy full SHA for 77842e6

• Implement a basic freeze immutable global tensors pass.

  Show description for c653d96
  

  ashwinmurthy

  authored and

  tensorflower-gardener

  committed
  c653d96

  Copy full SHA for c653d96

• Fix OSS windows paths for profiler.

  Show description for 2d84e07
  tensorflower-gardener

  committed
  2d84e07

  Copy full SHA for 2d84e07

• Add a canonicalize pattern for tf.Cast followed by tf.ReadVariableOp

  Show description for ff8da08
  

  ashwinmurthy

  authored and

  tensorflower-gardener

  committed
  ff8da08

  Copy full SHA for ff8da08

• Go: Update generated wrapper functions for TensorFlow ops.

  Show description for 00773c5
  tensorflower-gardener

  committed
  00773c5

  Copy full SHA for 00773c5

• Expose a c++ api for graph_def_to_tfl_flatbuffer for python converter to convert saved_model directly.

  Show description for ef95ed6
  

  renjie-liu

  authored and

  tensorflower-gardener

  committed
  ef95ed6

  Copy full SHA for ef95ed6

• Improve Conv2DBackpropInput to take input_sizes as a 2D shape.

  Show description for 68c2805
  

  Jingyue Wu

  authored and

  tensorflower-gardener

  committed
  68c2805

  Copy full SHA for 68c2805

Pagination

Previous

Next