298977887/ComfyUI
1
0
Fork 0
mirror of https://github.com/comfyanonymous/ComfyUI.git synced 2025-06-07 12:02:10 +08:00
Code Issues Packages Projects Releases Wiki Activity

throw away alpha channel in clip vision preprocessor (#7769)

Browse Source 
saves users having to explicitly discard the channel
This commit is contained in:
thot experiment 2025-04-23 18:28:36 -07:00 committed by GitHub
parent 11b68ebd22
commit e2eed9eb9b
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
comfy/clip_vision.py
View File

@ -18,6 +18,7 @@ class Output:
setattr(self, key, item) setattr(self, key, item)
def clip_preprocess(image, size=224, mean=[0.48145466, 0.4578275, 0.40821073], std=[0.26862954, 0.26130258, 0.27577711], crop=True): def clip_preprocess(image, size=224, mean=[0.48145466, 0.4578275, 0.40821073], std=[0.26862954, 0.26130258, 0.27577711], crop=True):
image = image[:, :, :, :3] if image.shape[3] > 3 else image
mean = torch.tensor(mean, device=image.device, dtype=image.dtype) mean = torch.tensor(mean, device=image.device, dtype=image.dtype)
std = torch.tensor(std, device=image.device, dtype=image.dtype) std = torch.tensor(std, device=image.device, dtype=image.dtype)
image = image.movedim(-1, 1) image = image.movedim(-1, 1)