dcnl_ml_predicting_detox_outcomes

Code used for the Machiune Learning Analyses in the Peer-Reviewed Article Titled: "Examining predictors of cocaine withdrawal syndrome at the end of detoxification treatment in women with cocaine use disorder".

DOI: 10.1016/j.jpsychires.2023.11.043

Summary of contents:

00_data_pre_processing.ipynb.ipynb - Code used for cleaning up the data, create outcomes based on CSSA split point and create downsampled dataset.

01.nested_cross_validation.ipynb - Nested cross validation process for 6 classification algorithms, figures (including SHAP), and results based on results for best algorithm.

02_make_datasets_for_others.ipynb - Creating datasets for collaborators to calculate stats and do their part of the analysis.

dependencies.txt - Text file containing Python and Python package versions used for this analysis.