OpenQDC: Open Quantum Data Commons

TL;DR

OpenQDC consolidates and standardizes a vast collection of quantum-mechanical datasets to accelerate the development of machine learning interatomic potentials for molecular dynamics simulations.

Contribution

It introduces the openQDC package, unifying 37 datasets with 400 million geometries, and provides tools for easy access, preprocessing, and benchmarking in MLIP research.

Findings

Benchmarking reveals challenges for existing architectures.

OpenQDC enables standardized MLIP training and evaluation.

The resource promotes collaboration and innovation in quantum chemistry and ML.

Abstract

Machine Learning Interatomic Potentials (MLIPs) are a highly promising alternative to force-fields for molecular dynamics (MD) simulations, offering precise and rapid energy and force calculations. However, Quantum-Mechanical (QM) datasets, crucial for MLIPs, are fragmented across various repositories, hindering accessibility and model development. We introduce the openQDC package, consolidating 37 QM datasets from over 250 quantum methods and 400 million geometries into a single, accessible resource. These datasets are meticulously preprocessed, and standardized for MLIP training, covering a wide range of chemical elements and interactions relevant in organic chemistry. OpenQDC includes tools for normalization and integration, easily accessible via Python. Experiments with well-known architectures like SchNet, TorchMD-Net, and DimeNet reveal challenges for those architectures and constitute a leaderboard to accelerate benchmarking and guide novel algorithms development. Continuously adding datasets to OpenQDC will democratize QM dataset access, foster more collaboration and innovation, enhance MLIP development, and support their adoption in the MD field.