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
[AOTI] Fix a weight loading issue when the weight size can be 0 #114280
Conversation
Summary: When a weight tensor is 0-size, no device memory should be allocated for it. This PR fixes the weight loading logic for such a case. This problem was found when running the 14K model test. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/114280
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 7bbb37d with merge base 54d0455 (): This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Summary: When a weight tensor is 0-size, no device memory should be allocated for it. This PR fixes the weight loading logic for such a case. This problem was found when running the 14K model test. ghstack-source-id: 852d91a2216a49361899396366559c2c6f85ad90 Pull Request resolved: #114280
… be 0" Summary: When a weight tensor is 0-size, no device memory should be allocated for it. This PR fixes the weight loading logic for such a case. This problem was found when running the 14K model test. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 aakhundov ColinPeppler [ghstack-poisoned]
Summary: When a weight tensor is 0-size, no device memory should be allocated for it. This PR fixes the weight loading logic for such a case. This problem was found when running the 14K model test. ghstack-source-id: cb49231f5be3f38a32ed187212bd92b43505657f Pull Request resolved: #114280
@pytorchbot merge |
Merge failedReason: This PR needs a If not, please add the To add a label, you can comment to pytorchbot, for example For more information, see Details for Dev Infra teamRaised by workflow job |
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
…rch#114280) Summary: When a weight tensor is 0-size, no device memory should be allocated for it. This PR fixes the weight loading logic for such a case. This problem was found when running the 14K model test. Pull Request resolved: pytorch#114280 Approved by: https://github.com/chenyang78
Stack from ghstack (oldest at bottom):
Summary: When a weight tensor is 0-size, no device memory should be allocated for it. This PR fixes the weight loading logic for such a case. This problem was found when running the 14K model test.
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @aakhundov @ColinPeppler