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

fix num_microbatches input for PP #781

Merged
merged 2 commits into from
Jan 7, 2025
Merged

Conversation

H-Huang
Copy link
Member

@H-Huang H-Huang commented Jan 7, 2025

Stack from ghstack (oldest at bottom):

PP is using the pipeline_parallel_degree as the default number of microbatches when it should be pipeline_parallel_degree * len(stages) for multi-stage schedules.

Also fixed up some logging / added logging.

H-Huang added a commit that referenced this pull request Jan 7, 2025
ghstack-source-id: 20fc75fbdcd6283459c4dd3b98df183ad32fe094
Pull Request resolved: #781
@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Meta Open Source bot. label Jan 7, 2025
@H-Huang H-Huang requested review from tianyu-l and wconstab and removed request for tianyu-l January 7, 2025 21:54
Copy link
Contributor

@tianyu-l tianyu-l left a comment

Choose a reason for hiding this comment

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

lgtm!

torchtitan/parallelisms/pipelining_utils.py Outdated Show resolved Hide resolved
PP is using the `pipeline_parallel_degree` as the default number of microbatches when it should be `pipeline_parallel_degree * len(stages)` for multi-stage schedules. 

Also fixed up some logging / added logging.


[ghstack-poisoned]
H-Huang added a commit that referenced this pull request Jan 7, 2025
ghstack-source-id: 459e8bee48fd77b027fecdc9e5f78ec375b87cb6
Pull Request resolved: #781
@H-Huang H-Huang merged commit 40d2aec into gh/H-Huang/18/base Jan 7, 2025
4 checks passed
H-Huang added a commit that referenced this pull request Jan 7, 2025
ghstack-source-id: 459e8bee48fd77b027fecdc9e5f78ec375b87cb6
Pull Request resolved: #781
@H-Huang H-Huang deleted the gh/H-Huang/18/head branch January 7, 2025 23:45
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Meta Open Source bot.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants