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

Fix dtype for MPS in reinforcement learning example #19982

Merged
merged 5 commits into from
Jun 21, 2024
Merged

Fix dtype for MPS in reinforcement learning example #19982

Changes from 3 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
c204de8
fix example for mps; issue #19981
swyo Jun 17, 2024
74f8a35
[pre-commit.ci] auto fixes from pre-commit.com hooks
pre-commit-ci[bot] Jun 17, 2024
7df6c44
Merge branch 'master' into fabric/example/rl-for-mps
swyo Jun 21, 2024
26006e7
apply feedback: type casting unconditionally
swyo Jun 21, 2024
7a62348
resolve conflict
swyo Jun 21, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion examples/fabric/reinforcement_learning/train_fabric.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -146,7 +146,9 @@ def main(args: argparse.Namespace):
# Single environment step
next_obs, reward, done, truncated, info = envs.step(action.cpu().numpy())
done = torch.logical_or(torch.tensor(done), torch.tensor(truncated))
rewards[step] = torch.tensor(reward, device=device).view(-1)
rewards[step] = torch.tensor(
reward, device=device, dtype=torch.float32 if device.type == "mps" else None
).view(-1)
awaelchli marked this conversation as resolved.
Show resolved Hide resolved
next_obs, next_done = torch.tensor(next_obs, device=device), done.to(device)

if "final_info" in info:
Expand Down
Loading