Skip to content

Reduce peak memory usage when freezing parameters. #14

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

Merged
merged 1 commit into from
Aug 16, 2022
Merged

Reduce peak memory usage when freezing parameters. #14

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion big_vision/optax.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -68,7 +68,7 @@ def create_schedule(mult=1.0, **kw):
# Removes weight decay updates. Note that weight decay already has an
# independent mask (which cannot be combined easily with a second mask),
# so instead we multiply updates for frozen params with zero.
optax.masked(optax.scale(0.0), frozen_mask)
optax.masked(optax.set_to_zero(), frozen_mask)
]

# Gradient clipping.
Expand Down