Commits

Breadcrumbs

History for

onv

User selector

All users

All time

Commit History

Commits on Feb 7, 2017

• Improve performance of ResizeArea op:

  Show description for 74be1df
  

  tensorflower-gardener

  authored and

  gunan

  committed
  74be1df

  Copy full SHA for 74be1df

• Add colocation constraints to StagingArea.

  Show description for a0087e2
  

  tensorflower-gardener

  authored and

  gunan

  committed
  a0087e2

  Copy full SHA for a0087e2

• Add a new lightweight queue-like object - RecordInput

  Show description for e99d3ca
  

  tensorflower-gardener

  authored and

  gunan

  committed
  e99d3ca

  Copy full SHA for e99d3ca

Commits on Feb 3, 2017

• Drop now-deleted ops.

  Show description for 4c6a924
  

  tensorflower-gardener

  authored and

  gunan

  committed
  4c6a924

  Copy full SHA for 4c6a924

Commits on Feb 1, 2017

• Restore deprecation warnings on {histogram,scalar}_summary in logging_ops.py (#7092)

  Show description for 08800ef
  

  wookayin

  authored and

  rmlarsen

  committed
  08800ef

  Copy full SHA for 08800ef

Commits on Jan 24, 2017

• Remove concat_v2 from r1.0 branch

  

  annarev

  authored and

  gunan

  committed
  ba45e40

  Copy full SHA for ba45e40

Commits on Jan 20, 2017

• Make "Neg" op hidden from Python.

  Show description for 510560e
  

  aselle

  authored and

  gunan

  committed
  510560e

  Copy full SHA for 510560e

• Change weight broadcasting for losses.sparse_softmax_cross_entropy to be the same as sparse_softmax_cross_entropy_with_logits. This means allowing greater than rank 2 inputs.

  Show description for 8a01c64
  

  tensorflower-gardener

  authored and

  gunan

  committed
  8a01c64

  Copy full SHA for 8a01c64

• Make dtype consistent with docs for TP/TN/FP/FN ops.

  Show description for 5572e35
  

  tensorflower-gardener

  authored and

  gunan

  committed
  5572e35

  Copy full SHA for 5572e35

• Restrict weights rank to be the same as the broadcast target, to avoid ambiguity on broadcast rules.

  Show description for b731e37
  

  tensorflower-gardener

  authored and

  gunan

  committed
  b731e37

  Copy full SHA for b731e37

Commits on Jan 17, 2017

• Merge commit for internal changes

  

  Patrick Nguyen

  committed
  19ff874

  Copy full SHA for 19ff874

Commits on Jan 15, 2017

• Add tf.tables_initializer as a replacement for tf.initialize_all_tables, and

  Show description for 9654cf9
  tensorflower-gardener

  committed
  9654cf9

  Copy full SHA for 9654cf9

• Default naming of scopes should continue, not reset, after jumps to reused scopes.

  Show description for e720678
  

  Lukasz Kaiser

  authored and

  tensorflower-gardener

  committed
  e720678

  Copy full SHA for e720678

Commits on Jan 14, 2017

• Automated rollback of change 142694447

  Show description for 7ab67d8
  tensorflower-gardener

  committed
  7ab67d8

  Copy full SHA for 7ab67d8

• Update einsum to check whether uncompacted_shape has multiple None values (#6830)

  Show description for ca09a89
  

  jihunchoi

  authored and

  drpngx

  committed
  ca09a89

  Copy full SHA for ca09a89

Commits on Jan 13, 2017

• Fix documentation sample code (#6836)

  Show description for 6b4dd84
  

  normanheckscher

  authored and

  drpngx

  committed
  6b4dd84

  Copy full SHA for 6b4dd84

• Makes the ops used by ParallelConcat internal.

  Show description for b24a0b2
  tensorflower-gardener

  committed
  b24a0b2

  Copy full SHA for b24a0b2

• Do parallel_stack as a graph rewrite instead of python code.

  Show description for 705cc93
  tensorflower-gardener

  committed
  705cc93

  Copy full SHA for 705cc93

• Make weight broadcasting more strict (in particular, catch the case where a 1-dim in values shape could be broadcast across a non-1 dimension in weights).

  Show description for 16cc887
  tensorflower-gardener

  committed
  16cc887

  Copy full SHA for 16cc887

• Merge changes from github.

  Show description for 769496b
  

  Patrick Nguyen

  authored and

  tensorflower-gardener

  committed
  769496b

  Copy full SHA for 769496b

• More new ops for speeding up input pipline: stage and unstage.

  Show description for 85acf52
  tensorflower-gardener

  committed
  85acf52

  Copy full SHA for 85acf52

• Create a benchmark for resize_bicubic_op.

  Show description for 221a8c6
  

  saeta

  authored and

  tensorflower-gardener

  committed
  221a8c6

  Copy full SHA for 221a8c6

• Remove unnecessary control dependencies.

  Show description for dac3cf8
  tensorflower-gardener

  committed
  dac3cf8

  Copy full SHA for dac3cf8

• Remove tf.mul, tf.neg and tf.sub from the public API

  Show description for 78b11ce
  

  aselle

  authored and

  tensorflower-gardener

  committed
  78b11ce

  Copy full SHA for 78b11ce

Commits on Jan 12, 2017

• Update docstring for SVD to emphasize that output argument order differs from numpy.linalg.svd.

  Show description for ef31cf3
  tensorflower-gardener

  committed
  ef31cf3

  Copy full SHA for ef31cf3

• Losses should support scalar logits/labels.

  Show description for 5ccc994
  tensorflower-gardener

  committed
  5ccc994

  Copy full SHA for 5ccc994

• Transpose confusion matrix output to make it consistent with sklearn.

  Show description for 4d9a240
  tensorflower-gardener

  committed
  4d9a240

  Copy full SHA for 4d9a240

• Fix documentation of rows vs columns of confusion matrix.

  Show description for 91685f8
  tensorflower-gardener

  committed
  91685f8

  Copy full SHA for 91685f8

Commits on Jan 11, 2017

• More conversions of flags library to argparse.

  Show description for 963674d
  

  Vijay Vasudevan

  authored and

  tensorflower-gardener

  committed
  963674d

  Copy full SHA for 963674d

• Removing TensorArray.pack and TensorArray.unpack

  Show description for a2466d1
  tensorflower-gardener

  committed
  a2466d1

  Copy full SHA for a2466d1

• Deprecate tf.neg, tf.mul, tf.sub (and remove math_ops.{neg,mul,sub} usages

  Show description for be270ec
  

  aselle

  authored and

  tensorflower-gardener

  committed
  be270ec

  Copy full SHA for be270ec

• Switch tf.concat_v2 references in third_party/tensorflow to tf.concat.

  Show description for 0e226af
  tensorflower-gardener

  committed
  0e226af

  Copy full SHA for 0e226af

• Automated rollback of change 141622306

  Show description for df52532
  

  Vijay Vasudevan

  authored and

  tensorflower-gardener

  committed
  df52532

  Copy full SHA for df52532

• Allow tf.divide's name argument to be honored.

  Show description for ca61311
  

  aselle

  authored and

  tensorflower-gardener

  committed
  ca61311

  Copy full SHA for ca61311

• Add tf.concat with the new argument order: tf.concat(values, axis, name). Also,

  Show description for 063801b
  tensorflower-gardener

  committed
  063801b

  Copy full SHA for 063801b