You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the example in the readme, the number of gt images must be equal to the number of mask images.
If you feel this is too troublesome, you can modify the logic in guided_diffusion/image_datasets.py.
For example:
I added the following code to this line, so that many gt images use the only mask image
# If there is only one mask image, reuse it
if len(mask_paths) == 1 and len(gt_paths) > 1:
mask_paths = [mask_paths[0]] * len(gt_paths)
what's wrong?
The text was updated successfully, but these errors were encountered: