Skip to content

set deterministic parameter in TrainModel#181

Merged
kathyxchen merged 4 commits intoFunctionLab:masterfrom
kathyxchen:change-seed-init
Oct 19, 2021
Merged

set deterministic parameter in TrainModel#181
kathyxchen merged 4 commits intoFunctionLab:masterfrom
kathyxchen:change-seed-init

Conversation

@kathyxchen
Copy link
Copy Markdown
Collaborator

Reference Issues/PRs

Fixes #180

What does this implement/fix? Explain your changes.

Moves

torch.backends.cudnn.deterministic = True
torch.backends.cudnn.benchmark = False

to TrainModel class

What testing did you do to verify the changes in this PR?

Running training and variant effect prediction to ensure the behavior is as expected

@kathyxchen kathyxchen merged commit 60e38a9 into FunctionLab:master Oct 19, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Slowdown of variant effect prediction with torch.backends.cudnn.deterministic = True

1 participant