Skip to content

fix(pt): replace in-place nlist masking to enable CUDA graph capture #5433

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
njzjz merged 1 commit into from
May 7, 2026
Merged

fix(pt): replace in-place nlist masking to enable CUDA graph capture #5433

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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: 2 additions & 2 deletions deepmd/pt/model/descriptor/repflows.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -496,8 +496,8 @@ def forward(
a_sw = a_sw.masked_fill(~a_nlist_mask, 0.0)
# set all padding positions to index of 0
# if the a neighbor is real or not is indicated by nlist_mask
nlist[nlist == -1] = 0
a_nlist[a_nlist == -1] = 0
nlist = torch.where(nlist == -1, 0, nlist)
a_nlist = torch.where(a_nlist == -1, 0, a_nlist)

# get node embedding
# [nframes, nloc, tebd_dim]
Expand Down
2 changes: 1 addition & 1 deletion deepmd/pt/model/descriptor/repformers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -457,7 +457,7 @@ def forward(

# set all padding positions to index of 0
# if the a neighbor is real or not is indicated by nlist_mask
nlist[nlist == -1] = 0
nlist = torch.where(nlist == -1, 0, nlist)
# nb x nall x ng1
if comm_dict is None:
assert mapping is not None
Expand Down
Loading