Commits

Branch selector

 master

User selector

zheng-xq

Datepicker

All time

Commit History

Commits on Sep 15, 2016

• Automated rollback of change 133188906

  Show description for 9bf15ad
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  9bf15ad

  Copy full SHA for 9bf15ad

Commits on Aug 27, 2016

• Adding Cudnn RNN support.

  Show description for 962dafe
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  962dafe

  Copy full SHA for 962dafe

Commits on Aug 19, 2016

• Add stream-executor changes to enable Cudnn fused LSTM/RNN support.

  Show description for 859e47f
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  859e47f

  Copy full SHA for 859e47f

Commits on Aug 5, 2016

• Merge pull request tensorflow#3668 from caisq/fix-cmake-1

  Show description for 32bd3d0
  zheng-xq

  authored
  32bd3d0

  Copy full SHA for 32bd3d0

• Merge changes from github.

  Show description for d89870d
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  d89870d

  Copy full SHA for d89870d

• Merge pull request tensorflow#3646 from PeterisP/libcupti-documentation

  Show description for f49afa0
  zheng-xq

  authored
  f49afa0

  Copy full SHA for f49afa0

• Merge pull request tensorflow#3656 from zheng-xq/branch_129393964

  Show description for c99b9be
  zheng-xq

  authored
  c99b9be

  Copy full SHA for c99b9be

• Merge pull request tensorflow#3615 from iver56/add-imports

  Show description for 2aec2b6
  zheng-xq

  authored
  2aec2b6

  Copy full SHA for 2aec2b6

• Merge commit for internal changes

  zheng-xq

  committed
  da41c02

  Copy full SHA for da41c02

Commits on Jul 11, 2016

• Increase the error threshold for a test.

  Show description for b45a6f2
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  b45a6f2

  Copy full SHA for b45a6f2

• Adding an environment variable to control whether to use Cudnn with AvgPool:

  Show description for 7f37ded
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  7f37ded

  Copy full SHA for 7f37ded

Commits on Jul 6, 2016

• Change the error level of Cuda host allocator memory allocation memory to

  Show description for 31ecbb0
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  31ecbb0

  Copy full SHA for 31ecbb0

Commits on Jul 1, 2016

• Clipping the negative padding for planes in conv3d.

  Show description for 1d9ada3
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  1d9ada3

  Copy full SHA for 1d9ada3

Commits on Jun 30, 2016

• Clip the padding for negative values.

  Show description for 78cf089
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  78cf089

  Copy full SHA for 78cf089

Commits on Jun 21, 2016

• Improve convolution autotune process. The max batch size VGG model can handle

  Show description for 466eb29
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  466eb29

  Copy full SHA for 466eb29

Commits on Jun 9, 2016

• Adding additional dependency to TensorFlow platform target.

  Show description for 8863a9c
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  8863a9c

  Copy full SHA for 8863a9c

Commits on Jun 2, 2016

• Add TF_EXTRA_CUDA_CAPABILITIES to support extra Cuda compute capabilities. The

  Show description for acc40ff
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  acc40ff

  Copy full SHA for acc40ff

Commits on May 26, 2016

• In case of memory allocation failure, return an error status to stream

  Show description for 8452d29
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  8452d29

  Copy full SHA for 8452d29

Commits on May 25, 2016

• Turn on CUDNN autotune by default.

  Show description for 19d33a1
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  19d33a1

  Copy full SHA for 19d33a1

Commits on May 24, 2016

• Change the ConvParameters comparison function to use std::tuple to avoid

  Show description for bf51a7c
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  bf51a7c

  Copy full SHA for bf51a7c

Commits on May 23, 2016

• Consider _HostSend and _HostRecv in allocation attribute inference. This makes

  Show description for d29fbf2
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  d29fbf2

  Copy full SHA for d29fbf2

Commits on May 16, 2016

• Add Autotune support for Winograd on Cudnn R5.

  Show description for 75b1f2f
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  75b1f2f

  Copy full SHA for 75b1f2f

Commits on May 11, 2016

• Add the autotune version for the backward passes. It is currently

  Show description for 939ede0
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  939ede0

  Copy full SHA for 939ede0

Commits on May 9, 2016

• Adding autotune to the Cudnn conv algorithm selection.

  Show description for 6ff265e
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  6ff265e

  Copy full SHA for 6ff265e

Commits on May 7, 2016

• Merge commit for internal changes

  zheng-xq

  committed
  230d2f9

  Copy full SHA for 230d2f9

Commits on May 6, 2016

• Merge pull request tensorflow#2250 from kashif/issue-2243

  Show description for 8b89c90
  zheng-xq

  committed
  8b89c90

  Copy full SHA for 8b89c90

• Remove duplicated dependency from BUILD.

  zheng-xq

  committed
  0eb6a46

  Copy full SHA for 0eb6a46

• Merge commit for internal changes

  zheng-xq

  committed
  46705e1

  Copy full SHA for 46705e1

Commits on Apr 29, 2016

• Adding tests for conv-grad when strides > ksize.

  zheng-xq

  committed
  8ec95eb

  Copy full SHA for 8ec95eb

Commits on Apr 27, 2016

• Issue 2066: Fix the conv for stride > ksize case.

  zheng-xq

  committed
  41a79f6

  Copy full SHA for 41a79f6

Commits on Apr 22, 2016

• Calling Cudnn calls in the current thread, instead of dispatching it into its

  Show description for b68860c
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  b68860c

  Copy full SHA for b68860c

Commits on Mar 25, 2016

• Change the order of notification before returning.

  Show description for 4735c21
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  4735c21

  Copy full SHA for 4735c21

Commits on Mar 22, 2016

• Fix range error from python3 on Mac.

  Show description for 5c4a648
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  5c4a648

  Copy full SHA for 5c4a648

Commits on Mar 18, 2016

• Change Cuda pinned memory allocation to BFC allocator.

  Show description for 2b7ca5c
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  2b7ca5c

  Copy full SHA for 2b7ca5c

Commits on Mar 10, 2016

• Improve the BiasGrad for NCHW using less shared memory and better memory

  Show description for 7cc6ba7
  

  zheng-xq

  authored and

  tensorflower-gardener

  committed
  7cc6ba7

  Copy full SHA for 7cc6ba7

Pagination

Previous

Next