PR #17187 Model post process for zero stage3 training

Model post process for zero stage3 training #17187

pengwa merged 10 commits into main from pengwa/zero_post_process

pengwa added training

pengwa requested a review from

askhade 2 years ago

pengwa requested a review from

ajindal1 2 years ago

pengwa requested a review from

baijumeswani 2 years ago

askhade commented on 2023-08-22

askhade dismissed these changes on 2023-08-22

Base automatically changed from pengwa/zero_offload to main 2 years ago

pengwa dismissed their stale review 2 years ago

The base branch was changed.

rebase main

4e59594d

pengwa force pushed from 86704cfa to 4e59594d 2 years ago

github-advanced-security commented on 2023-08-24

Make PyTorch and ORT both work with ZeRoOffloadSubscriber

5f5beec6

fixes

6aa0b4d5

refinement

08e83021

minors

a25856d0

add convergence debug switches

5e8330a3

comment debug log

57300d97

github-advanced-security commented on 2023-08-30

Merge branch 'main' of https://github.com/microsoft/onnxruntime into …

a4a85584

fix bug

f7941b27

askhade commented on 2023-09-18

refine one comment

dd021960

askhade approved these changes on 2023-09-21

pengwa merged 6b7bce5e into main 2 years ago

pengwa deleted the pengwa/zero_post_process branch 2 years ago

Reviewers

askhade

github-advanced-security

ajindal1

baijumeswani

Assignees

No one assigned

Labels

training

Milestone

No milestone

onnxruntime Model post process for zero stage3 training #17187 Merged

Model post process for zero stage3 training #17187

onnxruntime
Model post process for zero stage3 training
#17187

Merged