-
Notifications
You must be signed in to change notification settings - Fork 26.3k
[FSDP][BE] Better error msg for incorrect device for training #92027
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
Conversation
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/92027
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit a5c32f5: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
| cpu_device = torch.device("cpu") | ||
| if self._offload_params: | ||
| p_assert( | ||
| flat_param.device == cpu_device, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We simplify this assert message because it is an internal assert. Accidentally moving the model to GPU after FSDP constructor will be caught before in _check_flat_params_on_expected_device().
…ing" [ghstack-poisoned]
…ing" Closes #90541. [ghstack-poisoned]
…ing" Closes #90541. [ghstack-poisoned]
…ing" Closes #90541. [ghstack-poisoned]
ghstack-source-id: 021adfd Pull Request resolved: pytorch#92027
Stack from ghstack:
r's current device iscuda:r#92035 [FSDP][RFC] Enforce rankr's current device iscuda:ruse_orig_params=True#91767 [FSDP] Do not clean FQNs even foruse_orig_params=Truedevice_id+ CPU offload test #92031 [FSDP][BE] Improvedevice_id+ CPU offload testprefixed_param_names->fqnsfor consolidation #92028 [FSDP][BE] Renameprefixed_param_names->fqnsfor consolidationCloses #90541.