README.md

This repo contains the baselines and our proposed framework code, the latest version of the code will be organized and released soon. To run the code:

First:

Python

Python >= 3.6 (recommended >= 3.9).

Miniconda or Anaconda are recommended to create a virtual python environment.

pip install torch==1.10.0+cu111 torchvision==0.11.0+cu111 torchaudio==0.10.0 -f https://download.pytorch.org/whl/torch_stable.html

After ensuring that PyTorch is installed correctly, you can install other dependencies via:

pip install -r requirements.txt

Second:

• Download Raw Data

  You can download all the raw datasets at Google Drive and unzip them to datasets/raw_data/.

• Pre-process Data

  cd /path/to/your/project
  python scripts/data_preparation/${DATASET_NAME}/generate_training_data.py
  

  Replace ${DATASET_NAME} with one of PEMS-BAY, PEMS03, PEMS04, PEMS07, PEMS08, or any other supported dataset. The processed data will be placed in datasets/${DATASET_NAME}.

  Or you can pre-process all datasets by.

  cd /path/to/your/project
  bash scripts/data_preparation/all.sh
  

• Pre-train Model if you want to pretrain your own model, choose a base model path and using run.py to generate the model yourself. Or you can use our pretrained model in training_log as well.

python run.py -c examples/GWNet/GWNet_PEMS04.py --gpus '0'

save the PATH

• Using pretrained model in the framework and test the performance

python test.py --cfg "PATH/TO/COFIG" --ckpt "PATH/TO/MODEL" --gpus "0" --task "create_data_store" --dstore_dir "./data_store/MODEL"   
python run_index_build.py --dstore_dir "./data_store/MODEL/" 
python test.py --cfg "PATH/TO/COFIG" --ckpt "PATH/TO/MODEL" --gpus "0" --task "knn_test" --dstore_dir "./data_store/MODEL"