This repository contains files related to the data generation, training, and evaluation for MLIP models. The repository is organized as follows:
Go to each subfolder for a detailed description.
The generate-datasets/
directory contains scripts to generate datasets for MLIP model training.
The training-allegro/
directory provides datasets and sample configuration files for the training process of Allegro MLIP models.
The training-deepmd/
directory provides datasets and sample configuration files for the training process of DeePMD MLIP models.
The evaluate-allegro/
directory contains scripts to perform FES estimation with Allegro.
The evaluate-deepmd/
directory contains scripts to perform FES estimation with DeePMD.