Remove CUDNN_BATCHNORM_SPATIAL_PERSISTENT mode for accuracy #13844
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
soumith
reviewed
Nov 12, 2018
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
Contributor
|
CC @csarofeen |
…13844) Summary: Pull Request resolved: pytorch#13844 In S163230, we've found CuDNN 7 upgrade causes accuracy drop in training convolution network such as ResNeXt-101 (~0% accuracy), and video R(2+1)D (65 --> 63%). We've fixed this in Caffe2 D9601217, and we should do the same to ATen as well. Reviewed By: ezyang Differential Revision: D13025486 fbshipit-source-id: fe54180f0a97e4b759ad81b9b2bbffde4aa33047
xw285cornell
force-pushed
the
export-D13025486
branch
from
acea5d5 to
544098e
Compare
zdevito
pushed a commit
to zdevito/ATen
that referenced
this pull request
Nov 13, 2018
Summary: Pull Request resolved: pytorch/pytorch#13844 In S163230, we've found CuDNN 7 upgrade causes accuracy drop in training convolution network such as ResNeXt-101 (~0% accuracy), and video R(2+1)D (65 --> 63%). We've fixed this in Caffe2 D9601217, and we should do the same to ATen as well. Reviewed By: ezyang Differential Revision: D13025486 fbshipit-source-id: 04f4f0d9af6287b0400ca1842fb2cdac1f8cdb70
ezyang
added
module: cudnn
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
In S163230, we've found CuDNN 7 upgrade causes accuracy drop in training convolution network such as ResNeXt-101 (~0% accuracy), and video R(2+1)D (65 --> 63%).
We've fixed this in Caffe2 D9601217, and we should do the same to ATen as well.
Differential Revision: D13025486