search

yolain / ComfyUI-Easy-Use

In order to make it easier to use the ComfyUI, I have made some optimizations and integrations to some commonly used nodes.
GNU General Public License v3.0
1.04k stars 71 forks source link

Image DetailTransfer报错 #372

Open neverlandzl opened 2 months ago

neverlandzl commented 2 months ago

File "/root/autodl-tmp/ComfyUI/custom_nodes/ComfyUI-Easy-Use/py/image.py", line 977, in transfer new_image = torch.lerp(target_tensor, new_image, mask) RuntimeError: The size of tensor a (592) must match the size of tensor b (593) at non-singleton dimension 3

yolain commented 2 months ago

same as https://github.com/yolain/ComfyUI-Easy-Use/issues/241

neverlandzl commented 2 months ago

My target image and source image and mask size are the same. I solve it by making image width and height a multiple of 8.

dkqjhoifh commented 6 days ago

Even if the output image size of the VAE decoder is fixed, the problem persists even if all image sizes are unified to multiples of 8

yolain commented 6 days ago

Even if the output image size of the VAE decoder is fixed, the problem persists even if all image sizes are unified to multiples of 8

Currently it has been modified to automatically adjust, just update it.