cyclomon
commented
4 weeks ago In the configure argument, you can change the image size during training & inference.
--load_size is size for initial image loading, and --crop_size is size for random cropping form initially loaded images.
the default setting for load_size and crop_size during training is 286 and 256.
and for inference, both are set as 256.
You can change those parameters as you want, but the crop_size must be the number multiples of 8 as our generator network contains 3 downsampling layers.
Hello, Thanks for your excellent work. I have a question about the inference processing. When I inference, the size of the input images resized 256x256 at the result. I would like to ask that If I want to keep the size of the input image with the result, what should I do?
Thank you so much.