-
Notifications
You must be signed in to change notification settings - Fork 681
[rl] refactor model registery #2194
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
base: gh/wwwjn/5/base
Are you sure you want to change the base?
Conversation
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
| f"(model_cls={train_spec.model_cls.__name__}, flavor={model_flavor})" | ||
| ) | ||
| The plugin is loaded at runtime when needed (in generator.py), following | ||
| the pattern used in native_sampler.py. |
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.
where is native_sampler.py?
|
|
||
| This function registers TorchTitan models (e.g., Qwen3) with vLLM's ModelRegistry. | ||
| It's called automatically when the unified module is imported, or can be called | ||
| explicitly for manual registration. |
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.
why do we need the explicit way -- explicitly calling it would require importing in the first place, which would have already register it?
Stack from ghstack (oldest at bottom):