CubePerturbMCF

Introduction

This is the implementation of the model proposed in a proceeding paper of AAAI 2021, A Multi-step-ahead Markov Conditional Forward Model with Cube Perturbations for Extreme Weather Forecasting.

Requirement

torch==1.4.0
torchfile==0.1.0
torchtext==0.4.0
torchvision==0.5.0
numpy==1.17.2
tqdm==4.42.0
sklearn==0.23.1
jupyter==1.0.0
jupyter-client==5.3.4
jupyter-console==6.0.0
jupyter-core==4.6.0

Data Preparation

The ExtremeWeather dataset can be downloaded from here

Preprocessing

Use data_preprocess_optimal_area.ipynb to preprocess:

1. In the step "2 Read Data", change the path of the data downloaded from previous step.
2. Change the label_type in the step "3. Get labels of original image ready for optimal area and time finding" to your target extreme weather event.
3. Set the path of the output folder in the step "7. Saving Preprocessed Data"
4. Then, go through all the preprocessing steps for the experiments input.

Experiments

Use cnn_exp.py to conduct experiment:

1. Change the folder_name and validation_folder_name in the cnn_exp.py to your preprocessed data.
2. Modifiy the output path at the end of the code. Or you can directly use the default output path.

python3 cnn_exp.py {perturbation rate} {perturbation type} {cube size}