Skip to content

Conversation

@pacman100
Copy link
Contributor

What does this PR do?

  1. States an important detail when using fsdp_cpu_ram_efficient_loading.
  2. Adds a section with a mapping between FSDP sharding strategies and DeepSpeed ZeRO Stages.

@HuggingFaceDocBuilderDev

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@pacman100 pacman100 marked this pull request as ready for review February 9, 2024 09:28
Copy link
Member

@BenjaminBossan BenjaminBossan left a comment

Choose a reason for hiding this comment

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

Thanks for improving the FSDP docs. I'm not knowledgeable enough to fully review the correctness, but from what I can tell, this LGTM. Left a few comments, but they're only nits.

Copy link
Contributor

@muellerzr muellerzr left a comment

Choose a reason for hiding this comment

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

Thanks for doing this @pacman100!

@pacman100 pacman100 merged commit 86228e3 into huggingface:main Feb 9, 2024
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