FBGEMM icon indicating copy to clipboard operation
FBGEMM copied to clipboard

Published 20 hours ago • verified publisher icon pytorch

Optimize the cache fetch for forward split, pt. 2

Open q10 opened this issue 2 years ago • 2 comments

Summary: This follows up the work on D47778422 by plumbing the uvm_cache_stats argument passing up to the Python API level. local_uvm_cache_stats is now zeroed out before the prefetch step as opposed to after, to allow for the data to be passed into the forward step.

Differential Revision: D47809950

q10 avatar Jul 28 '23 18:07 q10

Deploy Preview for pytorch-fbgemm-docs canceled.

Name Link
Latest commit 3aaf989ae5c3a747acb334dd82f171cd1242fa35
Latest deploy log https://app.netlify.com/sites/pytorch-fbgemm-docs/deploys/64c40f24d6a20a00089a90eb

netlify[bot] avatar Jul 28 '23 18:07 netlify[bot]

This pull request was exported from Phabricator. Differential Revision: D47809950

facebook-github-bot avatar Jul 28 '23 18:07 facebook-github-bot