See here for more detailed information.

Using Anaconda 4.8.3:

• conda env create -f environment_minimal.yml
• conda activate denn_minimal
• python setup.py develop

The minimal installation includes the dependencies required to reproduce the results of experiments. Additional dependencies include:

• ray for hyperparameter tuning (ray_tune.py)
• plotly for parallel plots

Substitute {key} with the appropriate problem key (e.g. "exp", "sho", "nlo", etc.) and follow instructions for each method below. Note: to reproduce NAS results, specify {key} as "coo".

DEQGAN:

• python denn/experiments.py --pkey {key} --gan

L1 / L2 / Huber:

• Define PyTorch loss in denn/config/{key}.yaml under training.loss_fn (L1="L1Loss", L2="MSELoss", Huber="SmoothL1Loss")
• python denn/experiments.py --pkey {key}

RK4 / FD:

• python denn/traditional.py --pkey {key}