Skip to contents

Plot Calibration Curve

Source: R/calibration.R
plot_calibration.Rd

Creates a reliability diagram (calibration plot) showing the relationship between predicted probabilities and observed frequencies.

Usage

plot_calibration(
  calib_result,
  show_histogram = TRUE,
  by_cluster = FALSE,
  title = "Calibration Plot",
  ...
)

Arguments

calib_result

Result from compute_calibration.

show_histogram

Logical. If TRUE, show histogram of predictions below the calibration curve. Default: TRUE.

by_cluster

Logical. If TRUE, show separate curves for each cluster. If FALSE, aggregate across clusters. Default: FALSE.

title

Character. Plot title. Default: "Calibration Plot".

...

Additional arguments passed to ggplot2 functions.

Value

A ggplot2 object.

Details

A reliability diagram plots mean predicted probability (x-axis) against mean observed frequency (y-axis) for each bin. A perfectly calibrated model follows the diagonal line (y = x). Points above the diagonal indicate underconfidence; points below indicate overconfidence.

The shaded band around the diagonal shows +/- 0.1 deviation as a reference. The ECE value is displayed in the plot subtitle.

See also

compute_calibration

Examples

if (FALSE) { # \dontrun{
calib <- compute_calibration(fit$resp, z_true)

# Basic calibration plot
p <- plot_calibration(calib)
print(p)

# Show by cluster
p <- plot_calibration(calib, by_cluster = TRUE)
print(p)

# Save to file
ggsave("calibration.pdf", p, width = 6, height = 5)
} # }