Skip to contents

Helpers for computing metric data from the subset dataset

Usage

assert_compute_metric_data(r)

compute_metric_data(predictor, response)

Arguments

r

[data.frame()]

dataframe resulting from compute_metric_data

predictor

[numeric()]

The scores of the predictor

response

[factor()]

The response value

Value

[data.frame()]

With columns:

  • type [character()]: Metric name

  • y [numeric()]: Metric value

  • score, norm_score, norm_rank [numeric()]: Raw, normalized, and normalized rank predictor value per group.

With an attribute:

  • limits [list()]: With one entry per metric type. Each entry is a numeric(2) that contains the plotting limits for the metric.

Functions

  • assert_compute_metric_data(): Assert compute_metric result data.frame types are correct

  • compute_metric_data(): Compute metric analysis