Metatensor and metatomic: foundational libraries for interoperable atomistic machine learning

Abstract

Incorporation of machine learning (ML) techniques into atomic-scale modeling has proven to be an extremely effective strategy to improve the accuracy and reduce the computational cost of simulations. It also entails conceptual and practical challenges, as it involves combining very different mathematical foundations, as well as software ecosystems that are very well developed in their own right, but do not share many commonalities. To address these issues and facilitate the adoption of ML in atomistic simulations, we introduce two dedicated software libraries. The first one, metatensor, provides multi-platform and multi-language storage and manipulation of arrays with many potentially sparse indices, designed from the ground up for atomistic ML applications. By combining the actual values with metadata that describes their nature and that facilitates the handling of geometric information and gradients with respect to the atomic positions, metatensor provides a common framework to enable data sharing between ML software -- typically written in Python -- and established atomistic modeling tools -- typically written in Fortran, C or C++. The second library, metatomic, provides an interface to store an atomistic ML model and metadata about this model in a portable way, facilitating the implementation, training and distribution of models, and their use across different simulation packages. We showcase a growing ecosystem of tools, including low-level libraries, training utilities, and interfaces with existing software packages that demonstrate the effectiveness of metatensor and metatomic in bridging the gap between traditional simulation software and modern ML frameworks.

References in corpus (51)

• Array Programming with NumPy
• Canonical sampling through velocity-rescaling
• Advanced capabilities for materials modelling with Quantum ESPRESSO
• CP2K: An Electronic Structure and Molecular Dynamics Software Package -- Quickstep: Efficient and Accurate Electronic Structure Calculations
• PLUMED 2: New feathers for an old bird
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• On representing chemical environments
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• Recent developments in the PySCF program package
• PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments and Partial Charges
• DScribe: Library of Descriptors for Machine Learning in Materials Science
• Comparing molecules and solids across structural and alchemical space
• DeePMD-kit v2: A software package for Deep Potential models
• Competing quantum effects in the dynamics of a flexible water model
• A Fourth-Generation High-Dimensional Neural Network Potential with Accurate Electrostatics Including Non-local Charge Transfer
• Physics-inspired structural representations for molecules and materials
• Improved initial guess for minimum energy path calculations
• Nuclear quantum effects enter the mainstream
• Efficient stochastic thermostatting of path integral molecular dynamics
• Symmetry-Adapted Machine-Learning for Tensorial Properties of Atomistic Systems
• Chemical Shifts in Molecular Solids by Machine Learning
• A Transferable Machine-Learning Model of the Electron Density
• i-PI: A Python interface for ab initio path integral molecular dynamics simulations
• Molecular enhanced sampling with autoencoders: On-the-fly collective variable discovery and accelerated free energy landscape exploration
• Data-Driven Collective Variables for Enhanced Sampling
• Incorporating long-range physics in atomic-scale machine learning
• Learning the electronic density of states in condensed matter
• Feature Optimization for Atomistic Machine Learning Yields A Data-Driven Construction of the Periodic Table of the Elements
• SchNetPack 2.0: A neural network toolbox for atomistic machine learning
• Recursive evaluation and iterative contraction of $N$-body equivariant features
• From enhanced sampling to reaction profiles
• Accurate molecular dynamics and nuclear quantum effects at low cost by multiple steps in real and imaginary time: using density functional theory to accelerate wavefunction methods
• i-PI 3.0: a flexible and efficient framework for advanced atomistic simulations
• Recognizing Local and Global Structural Motifs at the Atomic Scale
• A Bayesian approach to NMR crystal structure determination
• Equivariant representations for molecular Hamiltonians and N-center atomic-scale properties
• Deep Learning Collective Variables from Transition Path Ensemble
• Efficient implementation of atom-density representations
• Analysis of the Results of Metadynamics Simulations by metadynminer and metadynminer3d
• Fast and flexible long-range models for atomistic machine learning
• Exploring the design space of machine-learning models for quantum chemistry with a fully differentiable framework
• IRA: A shape matching approach for recognition and comparison of generic atomic patterns
• Learning Collective Variables with Synthetic Data Augmentation through Physics-Inspired Geodesic Interpolation
• A deep learning model for chemical shieldings in molecular organic solids including anisotropy
• Prediction rigidities for data-driven chemistry
• Adaptive energy reference for machine-learning models of the electronic density of states
• Efficient Implementation of Gaussian Process Regression Accelerated Saddle Point Searches with Application to Molecular Reactions
• Reconstructions and Dynamics of $\beta$-Lithium Thiophosphate Surfaces