Principled Approaches for Extending Neural Architectures to Function Spaces for Operator Learning (Dataset)

We present the datasets for the paper "Principled Approaches for Extending Neural Architectures to Function Spaces for Operator Learning." In this paper, we identify and distill the key principles for constructing practical implementations of mappings between infinite-dimensional function spaces. Using these principles, we propose a recipe for converting several popular neural architectures into neural operators with minimal modifications.

Our paper includes experiments on a Navier-Stokes dataset at several resolutions: 64x64, 128x128, 256x256, 512x512 and 1024x1024. This repository contains the exact training/testing split that we used in training and evaluation for our experiments. The code in our GitHub repository preprocesses and constructs the training/testing data from the raw datasets above.