cvlab-columbia / pix2gestalt

Code for the paper "pix2gestalt: Amodal Segmentation by Synthesizing Wholes" (CVPR 2024)
https://gestalt.cs.columbia.edu/
Other
134 stars 10 forks source link

config.json file missing from HF #7

Closed lange4531 closed 4 months ago

lange4531 commented 4 months ago

Hi there!

When will you be releasing the config.json file?

egeozguroglu commented 4 months ago

Hi, our training and fine-tuning config was released here.

You don't need an additional config.json for Huggingface. Please follow our Gradio demo, which is self-sufficient for inference. Does this clarify? Feel free to follow up with any further questions.