-
Notifications
You must be signed in to change notification settings - Fork 229
[misc] Improve distributed related env variables and setup #487
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
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.
lets we just move this and related V1 preprocessing files (except those under scripts) to under fastvideo/v1?
| fastvideo_args = FastVideoArgs(model_path=args.model_path, | ||
| num_gpus=world_size, | ||
| device_str="cuda", | ||
| num_gpus=get_world_size(), |
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.
perhaps we should just remove num_gpus altogether, including in inference? If we always hardcode num_gpus=get_world_size() users would need to control it using CUDA_VISIBLE_DEVICES. When I added num_gpus I was only considering inference and also assumed tp_size == sp_size
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 think num_gpus is easier to control, if not specified we can use world size
|
Thanks! |
JerryZhou54
left a comment
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
Uh oh!
There was an error while loading. Please reload this page.