Real-Life Examples of Nice

Among other things, this repository contains scripts and notebooks to contextualize NICE into real-world problems. These examples are similar to the procedures reported in Jigyasa Nigam, Sergey Pozdnyakov, and Michele Ceriotti. “Recursive evaluation and iterative contraction of N-body equivariant features.” The Journal of Chemical Physics 153.12 (2020): 121101, but not direct productions.

In qm9_home_pc.ipynb and qm9_small.ipynb construct similar machine learned potentials for the QM9 dataset (see below). qm9_home_pc.ipynb is intended to run on a local workstation, whereas qm9_small.ipynb is best suited for HPC resources. We have also provided examples for the methane dataset (https://archive.materialscloud.org/record/2020.110). All notebooks include general advice on appropriate real-life hyperparameters.

QM9 dataset is available in the form of separate .xyz files for each molecule in such a special format that it can not be read by ase. The first cells of qm9_home_pc.ipynb and qm9_small.ipynb notebooks contain code to fetch the raw QM9 dataset and parses it into a single ase .extxyz file.