A Library for Learning Neural Operators

TL;DR

NeuralOperator is an open-source Python library that facilitates the training and deployment of neural operators, enabling generalization across function spaces with discretization-invariant properties, within the PyTorch ecosystem.

Contribution

The paper introduces NeuralOperator, a comprehensive, user-friendly library for neural operator learning that supports discretization-invariant training and inference, integrating cutting-edge models in an open-source package.

Findings

Supports training on various discretizations

Ensures discretization convergence properties

Provides a high-quality, easy-to-use interface

Abstract

We present NeuralOperator, an open-source Python library for operator learning. Neural operators generalize neural networks to maps between function spaces instead of finite-dimensional Euclidean spaces. They can be trained and inferenced on input and output functions given at various discretizations, satisfying a discretization convergence properties. Part of the official PyTorch Ecosystem, NeuralOperator provides all the tools for training and deploying neural operator models, as well as developing new ones, in a high-quality, tested, open-source package. It combines cutting-edge models and customizability with a gentle learning curve and simple user interface for newcomers.