### Installations
First, make the virtual environment
Second, install pytorch according to your computer environments.
Finally, install required packages
    ---- console -------
    conda create -n venv python=3.9
    conda activate venv
    pip   install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
    pip   install -r requirements.txt
    --------------------

## Experiments
You set arguments (model, coeff, seed, gpu) and run the following command,
    python run_bench.py model={model} coeff={coeff} seed={seed} gpu={gpu}
    
    arguments:
    mandatory
    (1) model: choose from {msm, rmsn, crn, gnet, ct, edct, edts} as baseline models
    (2) coeff: set the amount of time-varying coufonders (γ_w and γ_d in this paper) as an integer
    (3) seed : set the random seed as an integer 
    optional
    (4)  gpu : gpu number; If you omit this argument, gpu 0 will be used.
    
    Example:
    ---- console -------
    python run_bench.py model=msm  coeff=4 seed=10
    python run_bench.py model=crn  coeff=2 seed=101
    python run_bench.py model=edts coeff=0 seed=1010 gpu=1
    --------------------
   