Request: add CLIP Image-2-Prompt to the image classifier

There has been some significant improvement of object recognition and reverse prompt engineering for Stable Diffusion that could be ported over to RetroBatch. Specifically, having the ability to replace YOLO with OpenAI’s CLIP method would improve classification a great deal. There currently isn’t a smooth way to do this kind of NLP in a MacOS GUI.

Hello, and welcome.

Thanks for the info on the new models - I’ll see what I can do in a future release!