Thank you for the interest in the project!
Yeah it should be possible to change the method to work with grayscale images!
There are two approaches you can take:
Concat grayscale images 3x along the channel axis to make it into 3 channel output
You can change the last conv layer to have 1 output channel instead of 3 output channels
Is there a way to do rgb to grayscale image
Train_A would be RGB Train_B Would be grayscale
and want output to be grayscale