Code to load the model and run inference from HF?

[jkallini]

Hi, thank you for sharing your work on H-Nets! Very cool work!

It seems that the repository doesn’t currently include code or examples showing how to load the pretrained model from Hugging Face and run inference.

Could you please add:

• a short script or notebook demonstrating how to load the model (e.g., using transformers or similar API), and
• an example of running inference on sample input?

This would help users quickly test and reproduce results.
Thanks again for making your research available!

[malcolmsharpe]

If it helps, I posted a (unpolished) repo that does this using Modal for the GPU compute, and it wouldn't be hard to adapt for local use.