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.

Sign up for GitHub

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

Jump to bottom

CSC (class-specific context) CoOp/CoCoOp in "base-new" and "cross-dataset". #69

Open

galaxyzen opened this issue Nov 8, 2023 · 0 comments

galaxyzen commented Nov 8, 2023 •

edited

Loading

CSC (class-specific context) will generate context for every class:

  # CoOp.py
  # random initialization
  if cfg.TRAINER.COOP.CSC:
      print("Initializing class-specific contexts")
      ctx_vectors = torch.empty(n_cls, n_ctx, ctx_dim, dtype=dtype)

If I train the model with the setting cfg.TRAINER.COOP.CSC on a dataset which has 10 classes, can I eval this model on another dataset which may have 11 classes? I think the setting cfg.TRAINER.COOP.CSC causes the model is specific to n_cls .
If the cfg.TRAINER.COOP.CSC is turned on, I think the "base-new" experiment can't be conducted as well.

The text was updated successfully, but these errors were encountered:

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment