NeuralOperator/neuraloperator ? reverse-engineered prompt

Build me a Python library for learning neural operators in PyTorch, with the official Fourier Neural Operator models included and a few other operator learning architectures too.

I want it to make it easy to train on my own data, so someone should be able to import a model, set the input and output channels, choose the resolution, and start training without a lot of setup. Please include a version that uses tensor factorization to cut down the number of parameters, and make sure the models can run on data at different resolutions after training.

Also add a simple examples folder with a couple of practical demos, plus clear docs that explain what neural operators are and how to use them. If it helps, look up the current PyTorch and packaging docs online. I’d like the project to be installable locally in editable mode, and I want the code organized cleanly enough that people can extend it with new operator models later.