ChenRocks / ChenRocks/fast_abs_rl
raise RuntimeError("Mismatch in shape: grad_output[" RuntimeError: Mismatch in shape: grad_output[0] has a shape of torch.Size([1]) and output[0] has a shape of torch.Size([]).
Open
- Dominant language
- Python
- Stars
- 624
- Forks
- 183
- PR merge metrics
- No merged PRs in 30d
Description
deer
how to solve it？in the rl.py
autograd.backward(
[critic_loss] + losses,
[torch.ones(1).to(critic_loss.device)]*(1+len(losses)),
)
raise that problem
Contributor guide
No contributing guide indexed for this repository
Research direction
Start in rl.py at the autograd.backward call shown in the report and reproduce the RuntimeError. Inspect the shapes of critic_loss, losses, and the supplied gradients; done means the reported shape mismatch no longer occurs.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100