Skip to content

Conversation

@bddppq
Copy link
Contributor

@bddppq bddppq commented Apr 12, 2019

Stack:
    :white_circle:  #19633 Add is_mkldnn to at::Tensor  💚
    :white_circle:  #19204 Add aten mkldnn conv2d operator  💚
    :white_circle:  #19205 Add aten mkldnn ops: relu, max_pool2d and avg_pool2d  💚
    :white_circle:  #19206 Add aten mkldnn batch_norm operator  💚
    :black_circle:  #19207 Add aten mkldnn add operator  💚
    :white_circle:  #19209 Add aten mkldnn view operator  💚
    :white_circle:  #19210 Add aten mkldnn linear operator  💚
    :white_circle:  #19648 Adjust resnext run script  💚

Pull Request resolved: #19207

Differential Revision: D14889477

Differential Revision: D14889477
Differential Version: 79209394
bddppq added 5 commits April 12, 2019 21:15
Differential Revision: D14889477
Differential Version: 79296740
Differential Revision: D14889477
Differential Version: 79299279
Differential Revision: D14889477
Differential Version: 79689941
Differential Revision: D14889477
Differential Version: 79698403
Differential Revision: D14889477
Differential Version: 79729544
Differential Revision: D14889477
Differential Version: 80386614
Copy link
Contributor

@gchanan gchanan left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

approved once comments are addressed.

ideep::tensor& y = itensor_from_mkldnn(other);

ideep::tensor& z = itensor_from_mkldnn(result);
const std::vector<float> scales{1.0, alpha.to<float>()};
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

do we ever worry about mkldnn supporting double? We could technically write this generically with a dispatch macro.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ideep::sum::compute only takes scales as float vector

Differential Revision: D14889477
Differential Version: 80533697
bddppq added 4 commits April 23, 2019 14:17
Differential Revision: D14889477
Differential Version: 80534300
Differential Revision: D14889477
Differential Version: 80534911
Differential Revision: D14889477
Differential Version: 80541661
Differential Revision: D14889477
Differential Version: 80564052
bddppq added 2 commits April 23, 2019 21:42
Differential Revision: D14889477
Differential Version: 80580147
Differential Revision: D14889477
Differential Version: 80683615
bddppq added 4 commits April 25, 2019 18:20
Differential Revision: D14889477
Differential Version: 80773728
Differential Revision: D14889477
Differential Version: 80780050
Differential Revision: D14889477
Differential Version: 80785116
Differential Revision: D14889477
Differential Version: 80799725
zdevito pushed a commit to zdevito/ATen that referenced this pull request Apr 26, 2019
Summary: Pull Request resolved: pytorch/pytorch#19207

Reviewed By: dzhulgakov

Differential Revision: D14889477

fbshipit-source-id: 2c5e5ea5dfc26a9c9a172c5fa2c6d7584b167e16
@facebook-github-bot
Copy link
Contributor

This pull request has been merged in 61d5a8d.

zhangguanheng66 pushed a commit to zhangguanheng66/pytorch that referenced this pull request May 6, 2019
Summary: Pull Request resolved: pytorch#19207

Reviewed By: dzhulgakov

Differential Revision: D14889477

fbshipit-source-id: 2c5e5ea5dfc26a9c9a172c5fa2c6d7584b167e16
@ezyang ezyang deleted the export-D14889477 branch May 30, 2019 15:56
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

module: mkldnn Related to Intel IDEEP or oneDNN (a.k.a. mkldnn) integration

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants