-
-
Notifications
You must be signed in to change notification settings - Fork 906
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
[feature] add pytorch profiling #2182
Conversation
winglian
commented
Dec 13, 2024
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.
LGTM, but can we add inline comments and / or some documentation of this config option?
): | ||
if state.global_step == self.steps_to_profile: | ||
snapshot = torch.cuda.memory._snapshot() # pylint: disable=protected-access | ||
with open(Path(args.output_dir) / "snapshot.pickle", "wb") as fout: |
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.
Can we maybe output to a non-pickle file for safety reasons (shared to others , tampered, etc)? How would a person also read this file and display it?
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.
I don't believe there are any other options that the visualizer supports as input. I added docs but forgot to push them earlier.
* add pytorch profiling * kick off the profiler asap since things may get allcoated before train start * document feature * add url for visualizer [skip ci]
* add pytorch profiling * kick off the profiler asap since things may get allcoated before train start * document feature * add url for visualizer [skip ci]
* add pytorch profiling * kick off the profiler asap since things may get allcoated before train start * document feature * add url for visualizer [skip ci]