Usage

1. Dataset can be obtained from National Tibetan Plateau Data Center or dataset/dataset.csv.

2. Install Python 3.8. For convenience, execute the following command.

   pip install -r requirements.txt
   

3. Install pytorch_wavelets.

   git clone https://github.com/fbcotter/pytorch_wavelets
   
   cd pytorch_wavelets
   
   pip install .
   

4. Train and evaluate model. We provide the experiment scripts under the folder scripts/. You can reproduce the experiment results as the following examples:

   bash scripts/NSAF.sh