Add multi-tensor AdamW optimizer

facebookresearch / ClassyVision

An end-to-end PyTorch framework for image and video classification

https://classyvision.ai

MIT License

1.59k stars 278 forks source link

Closed renganxu closed 2 years ago

renganxu commented 2 years ago

Summary: Add the parallel version (multi-tensor) of AdamW for omnivore workflow, to improve the step and zero_grad performance in optimizer.

Reviewed By: mannatsingh

Differential Revision: D35660803

facebook-github-bot commented 2 years ago

This pull request was exported from Phabricator. Differential Revision: D35660803