Generate predictions from a climasus_ml model on new data

Applies the fitted XGBoost model to a new data frame. The new data must contain all feature columns used during training (see object$meta$feature_cols).

Usage

# S3 method for class 'climasus_ml'
predict(object, newdata, ...)

Arguments

object

A climasus_ml object from sus_mod_ml().

newdata

A data frame with the same feature columns as the training data. Missing features trigger an informative error.

...

Unused.

Value

A numeric vector of predictions (in the natural scale of the objective: counts for Poisson, values for squarederror, probabilities for binary logistic).