autoprognosis

AutoPrognosis: Automated Clinical Prognostic Modeling via Bayesian Optimization with Structured Kernel Learning

AutoPrognosis: A system for automating the design of predictive modeling pipelines tailored for clinical prognosis.

See <project_dir>/doc/install.md for installation instructions

Video Presentation Autoprognosis

You can find a presentation by Prof. van der Schaar describing AutoPrognosis here: https://www.youtube.com/watch?v=d1uEATa0qIo

Slides Autoprognosis

Usage

python3 autoprognosis.py -i <data.csv> --target <response variable> -o <outdir>  [ -n <num_sample> --it <num_iterations> ]

The results can be found in two json files: /result.json and report.json. They can be shown with:

python3 autoprognosis_report.py -i <outdir>

See also jupyter notebooks tutorial_autoprognosis_*.ipynb

Examples

mkdir result  # directory where the results generated by autoprognosis are stored
python3 autoprognosis.py -i ../../data/spambase.csv.gz --target label -o result --acquisitiontype MPI
python3 autoprognosis_report.py -i result --verbose 0  # --verbose 1 sorts by classifiers *and* parameters

Known issues

• Acquisition function LCB generates excesive warnings "The set cost function is ignored! LCB acquisition does not make sense with cost.". This can be ignored.

References

1. AutoPrognosis: Automated Clinical Prognostic Modeling via Bayesian Optimization with Structured Kernel Learning
2. Prognostication and Risk Factors for Cystic Fibrosis via Automated Machine Learning
3. Cardiovascular Disease Risk Prediction using Automated Machine Learning: A Prospective Study of 423,604 UK Biobank Participants