Rank Graduation Robustness

RGR measures how stable model predictions remain under perturbations.

Main functions

`safeai.rgr.rgr_score`(pred_original, ...[, ...])	Compute Rank Graduation Robustness between two prediction arrays.
`safeai.rgr.rgr_curve`(model, x_data, strengths, *)	Compute one Rank Graduation Robustness (RGR) curve and its AURGR value.
`safeai.rgr.aurgr_score`(model, x_data, ...)	Compute only the area under an RGR curve.
`safeai.rgr.compare_rgr`(models, strengths, ...)	Compare several models using Rank Graduation Robustness (RGR) curves.
`safeai.rgr.plot_rgr`(result, *[, model_name, ...])	Plot one RGR curve or a comparison of several RGR curves.