Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Jul 25, 2020

• Update SVDF tests to use more representative shapes and data. Restructure test to allow re-using float inputs outputs weights and biases in quantized kernels.

  Show description for 9d1863d
  

  Nat Jeffries

  authored and

  tensorflower-gardener

  committed
  9d1863d

  Copy full SHA for 9d1863d

• Go: Update generated wrapper functions for TensorFlow ops.

  Show description for 973cfa0
  tensorflower-gardener

  committed
  973cfa0

  Copy full SHA for 973cfa0

• Disable invalid rewrite Square(Sub(x, y)) => Identity(SquaredDiff(x,y)) for complex types.

  Show description for 10559ef
  tensorflower-gardener

  committed
  10559ef

  Copy full SHA for 10559ef

• Make StatusHelper safer to use and fix a related memory leak.

  Show description for 0de929f
  

  jiawenhao

  authored and

  tensorflower-gardener

  committed
  0de929f

  Copy full SHA for 0de929f

• [TF2XLA] [NFC] Simplify CaseOp::Compile, no need for an extra vector of pointers

  Show description for 4874f23
  

  cheshire

  authored and

  tensorflower-gardener

  committed
  4874f23

  Copy full SHA for 4874f23

• [TF2XLA] [NFC] Function are wonderful, let's use them (Simplifying IfOp::Compile)

  Show description for cd13583
  

  cheshire

  authored and

  tensorflower-gardener

  committed
  cd13583

  Copy full SHA for cd13583

• Remove private API usage of ragged_tensor.is_ragged.

  Show description for b2b9ca5
  

  haifeng-jin

  authored and

  tensorflower-gardener

  committed
  b2b9ca5

  Copy full SHA for b2b9ca5

• [Profiler] Add more metrics.

  Show description for 63dc8f8
  tensorflower-gardener

  committed
  63dc8f8

  Copy full SHA for 63dc8f8

• [tf.data] Adding support for serialization of input pipelines that use `from_tensors` or `from_tensor_slices` with dataset input.

  Show description for e8d0e18
  

  jsimsa

  authored and

  tensorflower-gardener

  committed
  e8d0e18

  Copy full SHA for e8d0e18

• Update reorder reshape with binary ops to handle activation functions too

  Show description for 6af4813
  

  karimnosseir

  authored and

  tensorflower-gardener

  committed
  6af4813

  Copy full SHA for 6af4813

• Refactor Python PIP build scripts for ARM/ARM64

  Show description for 98b4246
  

  terryheo

  authored and

  tensorflower-gardener

  committed
  98b4246

  Copy full SHA for 98b4246

• Disable eager_microbenchmarks_test for py38 oss build

  Show description for a1be543
  

  lindong28

  authored and

  tensorflower-gardener

  committed
  a1be543

  Copy full SHA for a1be543

• Add relevant build configurations for TPU compilation device

  Show description for 43eabe3
  

  Frank Chen

  authored and

  tensorflower-gardener

  committed
  43eabe3

  Copy full SHA for 43eabe3

• [tf.data] Adds a utility function that figures out how to rebatch a batch for any given worker.

  Show description for 6b6471f
  

  Rachel Lim

  authored and

  tensorflower-gardener

  committed
  6b6471f

  Copy full SHA for 6b6471f

• Hide tf.data Produce/Consume events in the trace viewer.

  Show description for d86293c
  

  Jiho Choi

  authored and

  tensorflower-gardener

  committed
  d86293c

  Copy full SHA for d86293c

• Fix zero point of -128 causing check error in tflite int8 add kernel.

  Show description for dd918be
  

  Nat Jeffries

  authored and

  tensorflower-gardener

  committed
  dd918be

  Copy full SHA for dd918be

• Merge pull request #41677 from yongtang:26069-fill-qint8-qint16-prerequisite

  Show description for 0770aec
  tensorflower-gardener

  committed
  0770aec

  Copy full SHA for 0770aec

Commits on Jul 24, 2020

• Pass non empty MLIR module serialized string when constructing TpuCompilationCacheKey.

  Show description for 3707838
  

  andyly

  authored and

  tensorflower-gardener

  committed
  3707838

  Copy full SHA for 3707838

• Merge pull request #41668 from vnvo2409:s3-get-children-provide-filesystem-support

  Show description for 5966258
  tensorflower-gardener

  committed
  5966258

  Copy full SHA for 5966258

• Update `run_deprecated_v1` tests with graph scope.

  Show description for cae7938
  

  yhliang2018

  authored and

  tensorflower-gardener

  committed
  cae7938

  Copy full SHA for cae7938

• Add release build to the kokoro continuous integration.

  Show description for 6a68d88
  

  advaitjain

  authored and

  tensorflower-gardener

  committed
  6a68d88

  Copy full SHA for 6a68d88

• Modified HLOAbsOp lowering for differing types.

  Show description for 8b62448
  

  rsuderman

  authored and

  tensorflower-gardener

  committed
  8b62448

  Copy full SHA for 8b62448

• Migrate Windows CPU official release and nightly testing builds to use the new bazel configs.

  Show description for 3528681
  

  Amit Patankar

  authored and

  tensorflower-gardener

  committed
  3528681

  Copy full SHA for 3528681

• Initial check-in of GPU XLA thunks dialect

  Show description for cf95331
  tensorflower-gardener

  committed
  cf95331

  Copy full SHA for cf95331

• Adds back a `DoCopy` in StridedSliceAssignOp that was accidentally deleted.

  Show description for 0943136
  

  wangpengmit

  authored and

  tensorflower-gardener

  committed
  0943136

  Copy full SHA for 0943136

• Only run RegisterTpuPlatform() statically when running on Google platforms

  Show description for 2a5e737
  

  Frank Chen

  authored and

  tensorflower-gardener

  committed
  2a5e737

  Copy full SHA for 2a5e737

• Fix 'unused variable' compiler error when building for ARM Cortex M4 in release mode

  Show description for dd9e169
  tensorflower-gardener

  committed
  dd9e169

  Copy full SHA for dd9e169

• Allow negative axes in tf.nn.sufficient_statistics even when shape of x is unknown.

  Show description for 926c086
  tensorflower-gardener

  committed
  926c086

  Copy full SHA for 926c086

• Update MultiWorkerMirroredStrategy API doc

  Show description for 87ab169
  

  crccw

  authored and

  tensorflower-gardener

  committed
  87ab169

  Copy full SHA for 87ab169

• Merge pull request #40954 from ROCmSoftwarePlatform:google_upstream_rocm_platform_fix_200630

  Show description for cbc87da
  tensorflower-gardener

  committed
  cbc87da

  Copy full SHA for cbc87da

• Change Tensorboard callback to directly write out the train_function graphs at the end of the first training batch, rather than trying to write the potentially-incorrect Keras backend graph when th…

  Show description for 4e5311d
  

  Tomer Kaftan

  authored and

  tensorflower-gardener

  committed
  4e5311d

  Copy full SHA for 4e5311d

• Merge pull request #41678 from grasskin:master

  Show description for e2204b2
  tensorflower-gardener

  committed
  e2204b2

  Copy full SHA for e2204b2

• Port the pack kernel to the new TfLiteEvalTensor API.

  Show description for beb4e6e
  

  Nick Kreeger

  authored and

  tensorflower-gardener

  committed
  beb4e6e

  Copy full SHA for beb4e6e

• [XLA] Fix a spurious verification failure when while has the same value passed in twice.

  Show description for c8be53e
  

  berkinilbeyi

  authored and

  tensorflower-gardener

  committed
  c8be53e

  Copy full SHA for c8be53e

• Added BatchMatMulV2 to the estimator.

  Show description for 42b7494
  tensorflower-gardener

  committed
  42b7494

  Copy full SHA for 42b7494

Pagination

Previous

Next