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

Consider using tensordict as underlying structure for subjects #1216

Open
fzimmermann89 opened this issue Sep 28, 2024 · 3 comments
Open

Consider using tensordict as underlying structure for subjects #1216

fzimmermann89 opened this issue Sep 28, 2024 · 3 comments
Labels
enhancement New feature or request question Further information is requested

Comments

@fzimmermann89
Copy link

Hi,

as a long-term goal, it might be worthwhile to consider changing Subject to inherit from tensordict instead of dict. This would allow batching of subjects and might solve some of the issues observed in #1179

The downside is, that to make use tensorsdicts functionality, some work is needed and compatibility to existing code might suffer.

Cheers,
Felix

@fzimmermann89 fzimmermann89 added the enhancement New feature or request label Sep 28, 2024
@fepegar
Copy link
Owner

fepegar commented Sep 29, 2024

Hi, @fzimmermann89. This sounds interesting.

This would allow batching of subjects

Can you please elaborate?

@fepegar fepegar added the question Further information is requested label Oct 5, 2024
@fepegar
Copy link
Owner

fepegar commented Oct 16, 2024

FYI I tried implementing Subject and Image as TensorDicts, but that didn't solve the collating issue.

@romainVala
Copy link
Contributor

too bad it looked promising

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request question Further information is requested
Projects
None yet
Development

No branches or pull requests

3 participants