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.

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] SAC Loss reshapes tensordict causing loss of time dimension #2338

Closed
matteobettini opened this issue Jul 30, 2024 · 0 comments · Fixed by #2494
Closed

[BUG] SAC Loss reshapes tensordict causing loss of time dimension #2338

matteobettini opened this issue Jul 30, 2024 · 0 comments · Fixed by #2494
Assignees
@kurtamohler
Labels
bug Something isn't working

Comments

@matteobettini
Copy link
Contributor

matteobettini commented Jul 30, 2024 •
edited
Loading

I am trying to use the SAC loss with a memory model.

The memory model expects data leading with batch dimension [B,T]

I pass data with this shape to the loss,
but here it gets reshaped

rl/torchrl/objectives/sac.py

Line 1150 in 371181c

tensordict_reshape = tensordict.reshape(-1)

causing the loss of the dimension T, which I cannot retrieve in my model.

Would it be possible to remove the reshaping of the data from the loss?
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@kurtamohler kurtamohler

Labels
bug Something isn't working
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

[BugFix] Avoid reshape(-1) for inputs to objectives modules kurtamohler/torchrl
3 participants
@kurtamohler @vmoens @matteobettini