Skip to content

skip resampling for randaffine(d) if possible #2264

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
wyli merged 4 commits into from
May 28, 2021
Merged

skip resampling for randaffine(d) if possible #2264

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
014aa33
skip resampling if possible
wyli May 27, 2021
e105416
update docstring
wyli May 28, 2021
dc7c695
Merge branch 'dev' into randaffine-optimise
wyli May 28, 2021
2261d6f
Merge branch 'dev' into randaffine-optimise
wyli May 28, 2021
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: 4 additions & 3 deletions monai/transforms/spatial/array.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1483,12 +1483,13 @@ def __call__(
See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample
"""
self.randomize()
# if not doing transform and spatial size not defined, nothing to do
# if not doing transform and spatial size doesn't change, nothing to do
# except convert to float and convert numpy/torch
if not self._do_transform and not spatial_size and not self.spatial_size:
sp_size = fall_back_tuple(spatial_size or self.spatial_size, img.shape[1:])
do_resampling = self._do_transform or (sp_size != ensure_tuple(img.shape[1:]))
if not do_resampling:
img = img.float() if isinstance(img, torch.Tensor) else img.astype("float32")
return torch.Tensor(img) if self.resampler.as_tensor_output else np.array(img)
sp_size = fall_back_tuple(spatial_size or self.spatial_size, img.shape[1:])
grid = self.get_identity_grid(sp_size)
if self._do_transform:
grid = self.rand_affine_grid(grid=grid)
Expand Down
2 changes: 1 addition & 1 deletion monai/transforms/spatial/dictionary.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -759,7 +759,7 @@ def __call__(

sp_size = fall_back_tuple(self.rand_affine.spatial_size, data[self.keys[0]].shape[1:])
# change image size or do random transform
do_resampling = self.rand_affine.spatial_size is not None or self._do_transform
do_resampling = self._do_transform or (sp_size != ensure_tuple(data[self.keys[0]].shape[1:]))

# to be consistent with the self._do_transform case (dtype and device)
affine = torch.as_tensor(np.eye(len(sp_size) + 1), device=self.rand_affine.rand_affine_grid.device)
Expand Down