Skip to content

add persistent rnns with conservative criteria#11248

Closed
ngimel wants to merge 2 commits intopytorch:masterfrom
ngimel:rnn_persist
Closed

add persistent rnns with conservative criteria#11248
ngimel wants to merge 2 commits intopytorch:masterfrom
ngimel:rnn_persist

Conversation

@ngimel
Copy link
Collaborator

@ngimel ngimel commented Sep 4, 2018

Persistent rnns provide much better performance on V100 with half input data for a variety of cases.

#if CUDNN_VERSION < 7200 || CUDA_VERSION < 9010
return CUDNN_RNN_ALGO_STANDARD;
#else
cudaDeviceProp* prop = at::cuda::getCurrentDeviceProperties();

This comment was marked as off-topic.

This comment was marked as off-topic.

Copy link
Contributor

@facebook-github-bot facebook-github-bot left a comment

Choose a reason for hiding this comment

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

ezyang is landing this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

zdevito pushed a commit to zdevito/ATen that referenced this pull request Sep 6, 2018
Summary:
Persistent rnns provide much better performance on V100 with half input data for a variety of cases.
Pull Request resolved: pytorch/pytorch#11248

Differential Revision: D9665687

Pulled By: ezyang

fbshipit-source-id: 2bd09a7eb1f5190aadb580977b0ba956e21a7dd5
PenghuiCheng pushed a commit to PenghuiCheng/pytorch that referenced this pull request Sep 11, 2018
Summary:
Persistent rnns provide much better performance on V100 with half input data for a variety of cases.
Pull Request resolved: pytorch#11248

Differential Revision: D9665687

Pulled By: ezyang

fbshipit-source-id: 2bd09a7eb1f5190aadb580977b0ba956e21a7dd5
@ngimel ngimel deleted the rnn_persist branch September 23, 2018 21:09
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants