Unet
Unet copied to clipboard
If the training images are in RGB channels
I'm glad to see this project. In my dataset, the images are in RGB channels (.tiff), what should i do to change the code adapting my dataset.
The label needs to be one channel grayscale image. It is not important whether the suffix is tiff or not. The input picture for training can be RGB. You only need to change all train-related tiffs in your program to the format of your picture.