Datasets:

sogeeking
/

wave_ood

train dict
{"bc":"dirichlet","c":512.5,"ic":"gaussian","k":57.5,"seed":0,"states":[[[[5.922798244759434e-25,5.9(...TRUNCATED)
{"bc":"dirichlet","c":512.5,"ic":"gaussian","k":57.5,"seed":1,"states":[[[[0.000012558822163555305,0(...TRUNCATED)
{"bc":"dirichlet","c":512.5,"ic":"gaussian","k":57.5,"seed":2,"states":[[[[1.4167075845783472e-10,1.(...TRUNCATED)
{"bc":"dirichlet","c":512.5,"ic":"gaussian","k":57.5,"seed":3,"states":[[[[1.3242581576128601e-38,1.(...TRUNCATED)
{"bc":"dirichlet","c":512.5,"ic":"gaussian","k":57.5,"seed":4,"states":[[[[1.7236207537497002e-28,1.(...TRUNCATED)
{"bc":"dirichlet","c":512.5,"ic":"gaussian","k":57.5,"seed":5,"states":[[[[3.6839478675992006e-33,3.(...TRUNCATED)
{"bc":"dirichlet","c":512.5,"ic":"gaussian","k":57.5,"seed":6,"states":[[[[5.5997432468757324e-8,5.5(...TRUNCATED)
{"bc":"dirichlet","c":512.5,"ic":"gaussian","k":57.5,"seed":7,"states":[[[[4.613185850617777e-26,4.6(...TRUNCATED)
{"bc":"dirichlet","c":512.5,"ic":"gaussian","k":57.5,"seed":8,"states":[[[[3.7874249869673804e-7,3.7(...TRUNCATED)
{"bc":"dirichlet","c":512.5,"ic":"gaussian","k":57.5,"seed":9,"states":[[[[0.00023191973741631955,0.(...TRUNCATED)

Out-Of-Distribution Evaluation Set For The 2D Wave Equation

Out-of-distribution evaluation set for the 2d wave equation. The underlying PDE is the wave equation.

This dataset contains out-of-distribution (OOD) samples for evaluating model generalization.

Data is stored in HDF5 (.h5) format.

This dataset was generated by Armand Kassai Koupaï and has been used in the following papers:

ZEBRA: In-Context Generative Pretraining for Solving Parametric PDEs — Louis Serrano, Armand Kassaï Koupaï, Thomas X Wang, Pierre Erbacher, Patrick Gallinari. ICML 2025. OpenReview
ENMA: Tokenwise Autoregression for Generative Neural PDE Operators — Armand Kassaï Koupaï, Lise Le Boudec, Louis Serrano, Patrick Gallinari. NeurIPS 2025.

See the full download script in the Zebra repository.

pip install huggingface_hub

python download_data/download_data_hugging_face.py --datasets wave_ood

import h5py

data = h5py.File("train.h5", "r")