README_en.md

Global Data-driven High-resolution Weather Model

English | 简体中文

This is a global data-driven high-resolution weather model implemented and improved by High-Flyer AI. It is the first AI weather model, which can compare with the ECMWF Integrated Forecasting System (IFS).

Typhoon track comparison:

Water vapour comparison:

Requirements

• hfai
• torch >=1.8

Training

The raw data is from the public dataset, ERA5 , which is integrated into the dataset warehouse, hfai.datasets.

1. pretrain

   submit the task to Yinghuo HPC:

    hfai python train/pretrain.py -- -n 8 -p 30

   run locally:

    python train/pretrain.py

2. finetune

   submit the task to Yinghuo HPC:

    hfai python train/fine_tune.py -- -n 8 -p 30

   run locally:

    python train/fine_tune.py

3. precipitation train

   submit the task to Yinghuo HPC:

    hfai python train/precipitation.py -- -n 8 -p 30

   run locally:

    python train/precipitation.py

Citation

@article{pathak2022fourcastnet,
  title={Fourcastnet: A global data-driven high-resolution weather model using adaptive fourier neural operators},
  author={Pathak, Jaideep and Subramanian, Shashank and Harrington, Peter and Raja, Sanjeev and Chattopadhyay, Ashesh and Mardani, Morteza and Kurth, Thorsten and Hall, David and Li, Zongyi and Azizzadenesheli, Kamyar and others},
  journal={arXiv preprint arXiv:2202.11214},
  year={2022}
}