Skip to content

Support variable number of channels in patch extraction #8

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

Open
LAguirr wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Support variable number of channels in patch extraction #8

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
5 changes: 4 additions & 1 deletion craft/craft_torch.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -171,7 +171,10 @@ def fit(self, inputs: np.ndarray):
strides = int(self.patch_size * 0.80)

patches = torch.nn.functional.unfold(inputs, kernel_size=self.patch_size, stride=strides)
patches = patches.transpose(1, 2).contiguous().view(-1, 3, self.patch_size, self.patch_size)

#Add num_channels to use black and white images
num_channels = inputs.shape[1]
patches = patches.transpose(1, 2).contiguous().view(-1, num_channels, self.patch_size, self.patch_size)

# encode the patches and obtain the activations
activations = _batch_inference(self.input_to_latent, patches, self.batch_size, image_size,
Expand Down