Content-Length: 257543 | pFad | http://github.com/pytorch/torchrec/pull/3025

8C Add new argument for sharding_plan by jd7-tr · Pull Request #3025 · pytorch/torchrec · GitHub
Skip to content

Add new argument for sharding_plan #3025

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

Closed
wants to merge 2 commits into from

Conversation

jd7-tr
Copy link
Contributor

@jd7-tr jd7-tr commented Jun 2, 2025

Summary: Add param to allow users to use their own sharding plan when calling shard_quant_model. This is helpful for users that do not rely on the TorchRec planner for the sharding plan.

Differential Revision: D75728320

jd7-tr added 2 commits June 1, 2025 21:32
Summary:

Rolling back the new `sharding_type` param. Other sharding types do not seem to work properly with the current `shard_quant_model` logic. Instead we are making changes to allow user to provide their own sharding plan for other sharding types (e.g. column-wise sharding).

Original commit changeset: bebefd06f38b

Original Phabricator Diff: D73540534

Differential Revision: D75727458
Summary: Add param to allow users to use their own sharding plan when calling `shard_quant_model`. This is helpful for users that do not rely on the TorchRec planner for the sharding plan.

Differential Revision: D75728320
@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jun 2, 2025
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D75728320

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 Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants








ApplySandwichStrip

pFad - (p)hone/(F)rame/(a)nonymizer/(d)eclutterfier!      Saves Data!


--- a PPN by Garber Painting Akron. With Image Size Reduction included!

Fetched URL: http://github.com/pytorch/torchrec/pull/3025

Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy