Internal function used by light_profile to calculate ALE profiles.
ale_profile(x, v, breaks = NULL, n_bins = 11, cut_type = c("equal",
"quantile"), value_name = "value", counts_name = "counts",
counts = TRUE, counts_weighted = FALSE, pred = NULL,
evaluate_at = NULL, indices = NULL, n_max = 1000, seed = NULL,
two_sided = FALSE, calibrate = TRUE)An object of class flashlight.
The variable to be profiled.
Cut breaks for a numeric v. Only used if no evaluate_at is specified.
Maxmium number of unique values to evaluate for numeric v. Only used if no evaluate_at is specified.
For the default "equal", bins of equal width are created for v by pretty. Choose "quantile" to create quantile bins.
Column name containing the profile value. Defaults to "value".
Name of the column containing counts if counts is TRUE.
Should counts be added?
If counts is TRUE: Should counts be weighted by the case weights? If TRUE, the sum of w is returned by group.
Optional vector with predictions.
Vector with values of v used to evaluate the profile. Only relevant for type = "partial dependence".
A vector of row numbers to consider.
Maximum number of ICE profiles to calculate within interval (not within data).
Integer random seed passed to light_ice.
Standard ALE profiles are calculated via left derivatives. Set to TRUE if two-sided derivatives should be calculated. Only works for continuous v. More specifically: Usually, local effects at value x are calculated using points between x-e and x. Set ale_two_sided = TRUE to use points between x-e/2 and x+e/2.
Should values be calibrated based on average preditions? Default is TRUE.
A tibble containing results.