Request for torchmd-train scripts for fine-tuning AceFF 2.0 model
#5
by
gcorrea96
- opened
Thank you for releasing AceFF 2.0 and for making the pretrained model available.
I would like to ask if it would be possible to share the torchmd-train training scripts used for AceFF 2.0. I am currently interested in fine-tuning the model on my own dataset and I would like to follow your pipeline as closely as possible to avoid implementation issues. If sharing the full training code is not possible, even a minimal example or documentation of the key training hyperparameters would be very helpful.
What's available publicly is in the GitHub of torchmd-net inside the scripts directory and there are examples there but not specific to aceff
https://github.com/torchmd/torchmd-net/
giadefa
changed discussion status to
closed