-
Notifications
You must be signed in to change notification settings - Fork 356
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
Several Improvements for the latest PyTorch Framework #1564
Open
MGAMZ
wants to merge
11
commits into
open-mmlab:main
Choose a base branch
from
MGAMZ:main
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
FutureWarning: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
FutureWarning: `torch.cuda.amp.GradScaler(args...)` is deprecated. Please use `torch.amp.GradScaler('cuda', args...)` instead.
MotivationWhen using torch.compile, the 'disable' key compile could override the compile. When disable is set to True, there should be no prompt like 'The model has been compiled'. ModificationAdd a simple check when compile is a dict, and check if there exists a key 'disable'. |
FSDP.optim_state_dict_to_load requires the following parameters: model: Module, optim: Optimizer, optim_state_dict: Dict[str, Any]
MGAMZ
changed the title
Fix Torch Future Warning on the deprecation of 'torch.cuda.amp'
Several Improvement for the latest PyTorch Framework
Aug 21, 2024
MGAMZ
changed the title
Several Improvement for the latest PyTorch Framework
Several Improvements for the latest PyTorch Framework
Aug 21, 2024
…tions The current runner implementation has not yet supported for pure-python style configurations on model wrapper class. I follow the mainstream implementation to support this feature.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Motivation
Fix torch Future Warning:
FutureWarning:
torch.cuda.amp.GradScaler(args...)
is deprecated. Please usetorch.amp.GradScaler('cuda', args...)
instead.Modification
Deprecate torch.cuda.amp and transfer to torch.amp, and partial it with
device='cuda'
. The modification will not introduce any influence on the following codes in this file