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.

Sign up for GitHub

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

Jump to bottom

mixed precision training support #4

Open

CarloLucibello opened this issue Feb 14, 2023 · 0 comments

Owner

CarloLucibello commented Feb 14, 2023 •

edited

Loading

Flux seems to support well f16 cast, so this should be doable.

Relevant resources

Mixed precision training paper
Some template code from @chengchingwen
Issue tracker for fp16 support in CUDA Tracker: Float16 support JuliaGPU/CUDA.jl#391
pytorch's GradScaler
discord discussion

The text was updated successfully, but these errors were encountered:

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment