jarvis.ai.uncertainty.gaussian_process_uncertainty

Code to predict properties and their uncertainty.

ML model used: lgbm

Module Contents

Functions

GaussianProcesses(x, y, jid[, cv, n_jobs, n_iter, ...])

Perform Gaussian Processes and determine prediction intervals.

jarvis.ai.uncertainty.gaussian_process_uncertainty.GaussianProcesses(x, y, jid, cv=2, n_jobs=-1, n_iter=10, random_state=508842607, scoring='neg_mean_absolute_error', prop='exfoliation_energy', write_files=True)

Perform Gaussian Processes and determine prediction intervals.

NOTE: it can deal with ~50 descriptors AT MAX ==> reduction of

the CFID descriptors is needed