Skip to content
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

SimpleHDFWriter extra seq lens not correct, not supporting custom seq lens #1671

Open
albertz opened this issue Jan 3, 2025 · 0 comments
Open

Comments

@albertz
Copy link
Member

albertz commented Jan 3, 2025

In SimpleHDFWriter.insert_batch, when using extra to put some other data (despite for the main data stream "data"), the seq lens are currently not handled correctly.

Currently the logic is:

        ...
        for i in range(n_batch):
            ...
                    for key, value in extra.items():
                        assert value.shape[0] == n_batch
                        self._insert_h5_other(key, value[i])

I.e. for some value of shape [B,T,F], if will use the seq len always as long as the max seq len of the current batch. That means:

  • The seq len is wrong.
  • There is no support for custom seq lens of extra in any case.

(cc @dorian-K @NeoLegends)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant