Skip to content

perf: skip bincount if unnecessary #4773

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
iProzd merged 2 commits into from
May 30, 2025
+14 −9
Merged

perf: skip bincount if unnecessary #4773

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
af9b980
perf: skip bincount if unnecessary
caic99 May 28, 2025
aefd3b5
fix ut
caic99 May 28, 2025
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
23 changes: 14 additions & 9 deletions deepmd/pt/model/network/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,17 +30,22 @@ def aggregate(
-------
output: [num_owner, feature_dim]
"""
bin_count = torch.bincount(owners)
bin_count = bin_count.where(bin_count != 0, bin_count.new_ones(1))

if (num_owner is not None) and (bin_count.shape[0] != num_owner):
difference = num_owner - bin_count.shape[0]
bin_count = torch.cat([bin_count, bin_count.new_ones(difference)])

# make sure this operation is done on the same device of data and owners
output = data.new_zeros([bin_count.shape[0], data.shape[1]])
if num_owner is None or average:
Comment thread
caic99 marked this conversation as resolved.
# requires bincount
bin_count = torch.bincount(owners)
bin_count = bin_count.where(bin_count != 0, bin_count.new_ones(1))
if (num_owner is not None) and (bin_count.shape[0] != num_owner):
difference = num_owner - bin_count.shape[0]
bin_count = torch.cat([bin_count, bin_count.new_ones(difference)])
else:
num_owner = bin_count.shape[0]
else:
bin_count = None

output = data.new_zeros([num_owner, data.shape[1]])
output = output.index_add_(0, owners, data)
if average:
assert bin_count is not None
output = (output.T / bin_count).T
return output

Expand Down