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

Jump to bottom

[Feature] dev-1.x change np.transpose to torch.permute for speed up #2277

Merged
merged 2 commits into from
Feb 18, 2023
Merged

[Feature] dev-1.x change np.transpose to torch.permute for speed up #2277

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5c50239
change np.transpose to torch.permute
sunjiahao1999 Feb 15, 2023
8b8cf9e
add comments
sunjiahao1999 Feb 15, 2023
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
12 changes: 8 additions & 4 deletions mmdet3d/datasets/transforms/formating.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -147,15 +147,19 @@ def pack_single_results(self, results: dict) -> dict:
if 'img' in results:
if isinstance(results['img'], list):
# process multiple imgs in single frame
imgs = [img.transpose(2, 0, 1) for img in results['img']]
imgs = np.ascontiguousarray(np.stack(imgs, axis=0))
imgs = [to_tensor(img) for img in results['img']]
ZwwWayne marked this conversation as resolved.
Show resolved Hide resolved
imgs = torch.stack(
imgs, dim=0).permute(0, 3, 1, 2).contiguous()
results['img'] = to_tensor(imgs)
else:
img = results['img']
if len(img.shape) < 3:
img = np.expand_dims(img, -1)
results['img'] = to_tensor(
np.ascontiguousarray(img.transpose(2, 0, 1)))
# To improve the computational speed by by 3-5 times, apply:
# `torch.permute()` rather than `np.transpose()`.
# Refer to https://github.com/open-mmlab/mmdetection/pull/9533
# for more details
results['img'] = to_tensor(img).permute(2, 0, 1).contiguous()

for key in [
'proposals', 'gt_bboxes', 'gt_bboxes_ignore', 'gt_labels',
Expand Down