Request for Source Code Related to Autoencoder for Gray-Scale Image Training

[zhangjie20231127]

I have been working with your excellent project,and I noticed that the pre-trained autoencoder model (autoencoder_vq_f4.pth) is designed for RGB images. However, I am currently working with gray-scale images and would like to train a new autoencoder model tailored for this task.

Could you please point me to the source code or provide guidance on how I can adapt the current autoencoder model for gray-scale image training?

[zsyOAOA]

Please refer to the following link:
https://github.com/CompVis/taming-transformers

[zhangjie20231127]

Thank you！