Commits

Branch selector

 c659697

User selector

All users

Datepicker

All time

Commit History

Commits on Dec 22, 2017

• [TF:XLA] Implement AdjustContrastv2, AdjustHue, AdjustSaturation, HSVToRGB, RGBToHSV in XLA bridge.

  Show description for c659697
  

  hawkinsp

  authored and

  tensorflower-gardener

  committed
  c659697

• Fixing tf.contrib.data API declarations.

  Show description for c0626e2
  

  jsimsa

  authored and

  tensorflower-gardener

  committed
  c0626e2

• [XLA] Remove Power architecture support from CPU backend.

  Show description for 2a6fb70
  

  hawkinsp

  authored and

  tensorflower-gardener

  committed
  2a6fb70

• In FirstReady node manager, use node name as tie-breaker when multiple nodes

  Show description for 87fcd47
  tensorflower-gardener

  committed
  87fcd47

• Fix allow_smaller_final_batches for bucket_by_sequence_length.

  Show description for 2f484f1
  

  Matthew Schulkind

  authored and

  tensorflower-gardener

  committed
  2f484f1

• Adds support in stream executor interface to update the scratch allocator used with a cuFFT plan. This enables plan reuse without requiring we keep the scratch allocation alive between executions.

  Show description for 4c0adf2
  tensorflower-gardener

  committed
  4c0adf2

• Extract placement restrictions for a graph

  Show description for dddbe5f
  

  benoitsteiner

  authored and

  tensorflower-gardener

  committed
  dddbe5f

• [XLA:GPU] Fix implementation of Thunk::ShouldBlockFutureThunks.

  Show description for 079a9b1
  

  Justin Lebar

  authored and

  tensorflower-gardener

  committed
  079a9b1

• No need to and don't add layout transform to control nodes.

  Show description for bb1b001
  

  Yao Zhang

  authored and

  tensorflower-gardener

  committed
  bb1b001

• Mark the Arg and RetVal ops as inexpensive.

  Show description for bb47c93
  

  mrry

  authored and

  tensorflower-gardener

  committed
  bb47c93

• Add comments to link convolutional factors code with definitions in the KFC paper.

  Show description for d1ca27b
  tensorflower-gardener

  committed
  d1ca27b

• Added an option to download a fresh release of clang for doing cuda_clang builds.

  Show description for fd29e95
  tensorflower-gardener

  committed
  fd29e95

• [XLA] Support conditional in all backends.

  Show description for 531bf79
  tensorflower-gardener

  committed
  531bf79

• Support multiple loss and multiple optimizers in replicate_model_fn.

  Show description for a2039aa
  

  isaprykin

  authored and

  tensorflower-gardener

  committed
  a2039aa

• Made hard colocation constraints (i.e constraints that must be met for the

  Show description for f1a1bd8
  

  benoitsteiner

  authored and

  tensorflower-gardener

  committed
  f1a1bd8

• [XLA] Print out missing extra-info for many instructions in the HLO graph dumper.

  Show description for b2aa695
  

  Justin Lebar

  authored and

  tensorflower-gardener

  committed
  b2aa695

• [XLA] Update the parsing and stringfication of the Dot dimension numbers.

  Show description for 584b48c
  tensorflower-gardener

  committed
  584b48c

• [XLA] Add zero-sized batchnorm test.

  Show description for 1654f7a
  

  Justin Lebar

  authored and

  tensorflower-gardener

  committed
  1654f7a

• Support fill op.

  Show description for 4b66819
  

  Yao Zhang

  authored and

  tensorflower-gardener

  committed
  4b66819

• [XLA] Run ZeroSizedHloElimination after BatchNormExpander.

  Show description for 2f77d96
  

  Justin Lebar

  authored and

  tensorflower-gardener

  committed
  2f77d96

• [XLA:GPU] Implement BatchNormThunk as a call into cudnn.

  Show description for 6a9a9ed
  

  Justin Lebar

  authored and

  tensorflower-gardener

  committed
  6a9a9ed

• Automated g4 rollback of changelist 179872807

  Show description for e909415
  

  annarev

  authored and

  tensorflower-gardener

  committed
  e909415

• No external change.

  Show description for 088bf69
  

  Dandelion Man?

  authored and

  tensorflower-gardener

  committed
  088bf69

• Internal change.

  Show description for 4274682
  

  annarev

  authored and

  tensorflower-gardener

  committed
  4274682

• Adding a test that checks that Grappler is able to infer shapes for core tf.data API.

  Show description for ad82d25
  

  jsimsa

  authored and

  tensorflower-gardener

  committed
  ad82d25

• Added a comment to explain why we can't reorder BiasAdd

  Show description for cd1add8
  

  benoitsteiner

  authored and

  tensorflower-gardener

  committed
  cd1add8

• Add test coverage for example parsing with a feature that is an empty string.

  Show description for 6b8b9e4
  

  mrry

  authored and

  tensorflower-gardener

  committed
  6b8b9e4

• Fold FuseBatchNorm correctly.

  Show description for e38cf4c
  

  Suharsh Sivakumar

  authored and

  tensorflower-gardener

  committed
  e38cf4c

• Extract placement restrictions for a graph

  Show description for af0e847
  

  benoitsteiner

  authored and

  tensorflower-gardener

  committed
  af0e847

• Update ops-related pbtxt files.

  Show description for 4c76bb4
  tensorflower-gardener

  committed
  4c76bb4

• Automated g4 rollback of changelist 179858783

  Show description for c9f0c0b
  

  blakehechtman

  authored and

  tensorflower-gardener

  committed
  c9f0c0b

• [XLA] Transform division-by-fp-constant to multiplication by its inverse.

  Show description for 0ea9532
  

  Justin Lebar

  authored and

  tensorflower-gardener

  committed
  0ea9532

Commits on Dec 21, 2017

• Internal change.

  Show description for 22a923d
  

  annarev

  authored and

  tensorflower-gardener

  committed
  22a923d

• Only squeeze the last dimension of outputs and indices in infer_graph of

  Show description for e528973
  tensorflower-gardener

  committed
  e528973

• Saving a couple of us in IteratorGetNextOp::ComputeAsync.

  Show description for 83b4195
  

  jsimsa

  authored and

  tensorflower-gardener

  committed
  83b4195

Pagination

PreviousNext