Skip to contents

Predict class labels or estimate conditional probabilities for the specified new data.

Usage

# S3 method for supclass
predict(
  object,
  newx,
  type = c("class", "probability"),
  selection = c("cv_1se", "cv_min", "all"),
  ...
)

Arguments

object

An object of class abclass.

newx

A numeric matrix representing the design matrix for predictions.

type

A character value specifying the desired type of predictions. The available options are "class" for predicted labels and "probability" for class conditional probability estimates.

selection

An integer vector for the solution indices or a character value specifying how to select a particular set of coefficient estimates from the entire solution path for prediction. If the specified object contains the cross-validation results, one may set selection to "cv_min" (or "cv_1se") for using the estimates giving the smallest cross-validation error (or the set of estimates resulted from the largest lambda within one standard error of the smallest cross-validation error) or prediction. The prediction for the entire solution path will be returned in a list if selection = "all" or no cross-validation results are available in the specified object.

...

Other arguments not used now.

Value

A vector representing the predictions or a list containing the predictions for each set of estimates.

Examples

## see examples of `supclass()`.