krasserm / super-resolution

Tensorflow 2.x based implementation of EDSR, WDSR and SRGAN for single image super-resolution
Apache License 2.0
1.5k stars 352 forks source link

NOOB Question: Input 0 is incompatible with layer model_3: #74

Open peoplecure opened 3 years ago

peoplecure commented 3 years ago

Hi, I have a noob question. I get an error message each time I use my image.

ValueError: Input 0 is incompatible with layer model_3: expected shape=(None, None, None, 3), found shape=(1, 499, 552, 4)

MY guess is that I need to remove the first three values in shape=(1, 499, 552, 4). Am I correct?

krasserm commented 3 years ago

Seems you're using an RGBA image (4 channels). You need to convert it to an RGB image first (3 channels).