Advanced Cortex Usage

For the full cortex.run() parameter reference — including caching, saving results, and batch processing — see Running Cortex.

Selective Processing

Use a CSV file to run specific participant-feature combinations instead of computing every feature for every participant:

cortex.run(
    study_id,
    features=['hometime', 'screen_duration'],
    run_part_and_feats="subset.csv"
)

The CSV should have participant and feature columns:

participant,feature
U1234567890,hometime
U1234567890,screen_duration
U0987654321,hometime

This is useful when reprocessing failed runs or targeting specific participants without recomputing the entire study.