files of interest:
    -synthetics: train/synthetics/single.py
    -pseudo-synthetic: train/synthetics/ps.py
    -real: train/real/real.py
    brain data files: /data/brain_data

install relevant modules and run with a simply: python file_of_interest.py
Attempt was made to configure path to brain data. May need to tweak.
Most files are configures to work with Wieghts and Biases framework (wandb). If you do not have an account, simply comment out the initialization with wanbd and paste a non-wandb setup outside.


PYTHON ENVIRONMENT SETUP

# for GDN model training
conda install pytorch torchvision torchaudio -c pytorch
conda install -c conda-forge pytorch-lightning
pip install --upgrade wandb
conda install -c conda-forge pytorch-lightning
conda install -c conda-forge tensorflow
conda install -c anaconda scipy
conda install -c anaconda networkx
conda install -c anaconda pillow
conda install -c conda-forge matplotlib
conda install -c conda-forge scikit-learn
conda install numba


# spectral templates environment: DO THIS - cvxpy doesnt play nice with pytorch
conda create -n st-env python=3.7.9
conda activate st-env
conda install -c conda-forge pytorch-lightning
pip install --upgrade wandb
conda install cvxpy # try without forge
conda install -c mosek mosek
conda install numba matplotlib networkx
conda install -c conda-forge scikit-learn

if Spectral Templates, which uses CVX and mosek, place mosek license in the home folder. Can be requested on mosek.com

# glasso, network deconvolution,  environment
conda create -n glasso-env python=3.7.9
conda activate glasso-env
conda install pytorch-lightning
pip install --upgrade wandb
conda install -c conda-forge scikit-learn
conda install numba matplotlib networkx



