New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
cudnn.deterministic mode #270
Comments
I would argue that if you want absolutely deterministic results, use |
@jpuigcerver cudnn has deterministic algorithms, see the manual |
Copy & paste from the cuDNN 5.1 user guide:
So, for instance, for training, using |
@jpuigcerver that's why I mentioned that we'd have to fall back to THNN or assert in MaxPooling. I think determinism within one version is good enough to have already. |
@jpuigcerver Just to be clear, MaxPooling is nondeterministic only when pooling stride is less than pooling window. |
I see that it's not clear sometimes how to enable deterministic mode, should we maybe instead of
setMode
have a global booleancudnn.deterministic
likecudnn.benchmark
orcudnn.fastest
? Would work with functional too.In case of MaxPooling we'd have to fall back to THNN, as far as I remember cudnn maxpooling is not deterministic.
The text was updated successfully, but these errors were encountered: