# Datasets of GAOT

This repository contains the datasets used in the paper  
**"Geometry Aware Operator Transformer as an Efficient and Accurate Neural Surrogate for PDEs on Arbitrary Domains."**

The repository includes two subfolders:

- `time_dep/` — time-dependent datasets  
- `time_indep/` — time-independent datasets  

In addition, some datasets used in the paper are provided by [RIGNO](https://arxiv.org/pdf/2501.19205).  
These can be downloaded from [Zenodo](https://zenodo.org/records/14765453).

For a detailed description of all datasets, please refer to **Supplementary Material, Table D.1** of the paper.

---

### Data Structure

Each file contains either two or three groups:

- **u**: the vector-valued solution function  
- **c**: the vector-valued known spatial parameters (constant in time)  
- **x**: the spatial coordinates in all directions (e.g., *x* and *y*)  

All arrays are 4-dimensional with the following axis order: `[batch, time, space, variables]`.