Skip to content

BUG: torch.arange: workaround for missing dtype implementations #405

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

Merged
ev-br merged 4 commits into from
Mar 17, 2026
Merged

BUG: torch.arange: workaround for missing dtype implementations #405

Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
4138c05
BUG: torch.arange: workaround for missing dtype implementations
lucascolley Mar 4, 2026
6fa8ef8
remove `empty` guard [skip ci]
lucascolley Mar 17, 2026
662d3ae
Update array_api_compat/torch/_aliases.py
lucascolley Mar 17, 2026
6a948ff
Update array_api_compat/torch/_aliases.py
ev-br Mar 17, 2026
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
7 changes: 5 additions & 2 deletions array_api_compat/torch/_aliases.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -616,8 +616,11 @@ def arange(start: float,
dtype = torch.int64
else:
dtype = torch.float32
return torch.empty(0, dtype=dtype, device=device, **kwargs)
return torch.arange(start, stop, step, dtype=dtype, device=device, **kwargs)
return torch.empty(0, device=device, **kwargs).to(dtype)
try:
return torch.arange(start, stop, step, dtype=dtype, device=device, **kwargs)
except (NotImplementedError, RuntimeError):
Comment thread
ev-br marked this conversation as resolved.
return torch.arange(start, stop, step, device=device, **kwargs).to(dtype)

# torch.eye does not accept None as a default for the second argument and
# doesn't support off-diagonals (https://github.com/pytorch/pytorch/issues/70910)
Expand Down
Loading