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

[BUG] Unbind resets nested batch sizes #469

Closed

matteobettini opened this issue Jul 3, 2023 · 1 comment · Fixed by #471

Assignees

Labels

bug

Contributor

matteobettini commented Jul 3, 2023

td = TensorDict({"a": TensorDict({"b": torch.zeros(2,3)}, [2,3])},[2])
td["a"].batch_size # torch.Size([2, 3])
tds = td.unbind(0)
tds[0]["a"].batch_size # torch.Size([]) should be torch.Size([3])

matteobettini added the bug label

matteobettini assigned vmoens

Contributor Author

matteobettini commented Jul 3, 2023 •

edited

Loading

I think this is a problem stemming from the fact that apply disregards the nested batch sizes. If you want I can look into it, just let me know

vmoens mentioned this issue

[Benchmark] Benchmark unbind #470

Merged

vmoens linked a pull request

that will close this issue

[BugFix] Fix unbind #471

Merged

matteobettini mentioned this issue

[BugFix] Fix unbind #471

Merged

vmoens closed this as completed in #471

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