Skip to content
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

Patch Speed Monitor MFU #2013

Merged

Conversation

mvpatel2000
Copy link
Contributor

@mvpatel2000 mvpatel2000 commented Feb 28, 2023

What does this PR do?

As title. Fixes speed monitor bug which has a problem with flops/MFU. We now keep a record of flop utilization, as the values might be different depending on the level of padding in the batch. We also correctly sum the flop counts from all the devices

image

What issue(s) does this change relate to?

CO-1838

Copy link
Contributor

@vchiley vchiley left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm

(with one nit cmt)

composer/callbacks/speed_monitor.py Show resolved Hide resolved
@mvpatel2000 mvpatel2000 merged commit 82a9e6d into mosaicml:dev Mar 1, 2023
@mvpatel2000 mvpatel2000 deleted the mvpatel2000/patch-speed-monitor-2 branch March 1, 2023 02:39
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.

4 participants