FuXi-CFD Dataset and Pre-trained Model (v1.0)
Abstract
This record provides the official archived metadata for the FuXi-CFD dataset and pre-trained model associated with the paper: Lin, C., et al. *Reconstructing fine-scale 3D wind fields with terrain-informed machine learning*, Nature Communications (2026). Version: v1.0 Total dataset size: ~250 GB Total cases: ~12,000 Due to the large dataset size, the full data and model weights are hosted externally on HuggingFace. Dataset repository: https://huggingface.co/datasets/linchensen/FuXi-CFD-dataset Model repository: https://huggingface.co/linchensen/FuXi-CFD-model This Zenodo record provides: - Persistent DOI for citation - Versioned metadata - Documentation snapshot - Long-term archival reference All data are released under CC BY-NC 4.0. For detailed data structure, variable definitions, and inference instructions, please refer to the corresponding HuggingFace repositories.
Fields of Study & Subjects
Meteorological Phenomena and Simulations
Wind and Air Flow Studies
Model Reduction and Neural Networks
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 25 February 2026
- Language
- EN
- Document type
- Dataset