Bins predicted probabilities and computes observed frequencies within
each bin. Useful for calibration plots of binary classifiers.
Usage
calibration_df(truth, prob, n_bins = 10L)
Arguments
- truth
numeric vector (0/1 or TRUE/FALSE).
- prob
numeric vector of predicted probabilities.
- n_bins
integer, number of bins (default 10).
Value
A data.frame with columns bin_mid, obs_freq,
mean_pred, n.
Examples
calibration_df(
truth = c(0, 0, 1, 1),
prob = c(0.1, 0.3, 0.7, 0.9),
n_bins = 2
)
#> bin_mid obs_freq mean_pred n
#> 1 0.25 0 0.2 2
#> 2 0.75 1 0.8 2
