predict.brif {brif}

Make predictions using a brif model

Description

Make predictions for newdata using a brif model object.

Usage

## S3 method for class 'brif'
predict(
  object,
  newdata = NULL,
  type = c("score", "class"),
  vote_method = 1,
  nthreads = 2,
  ...
)

Arguments

object

an object of class "brif" as returned by the brif training function.

newdata

a data frame. The predictor column names and data types must match those supplied for training. The order of the predictor columns does not matter though.

type

a character string indicating the return content. For a classification problem, "score" means the by-class probabilities and "class" means the class labels (i.e., the target variable levels). For regression, the predicted values are returned.

vote_method

an integer (0 or 1) specifying the voting method in prediction. 0: each leaf contributes the raw count and an average is taken on the sum over all leaves; 1: each leaf contributes an intra-node fraction which is then averaged over all leaves with equal weight.

nthreads

an integer specifying the number of threads used by the program. This parameter only takes effect on systems supporting OpenMP.

...

additional arguments.

Details

Note: If a model is built just for making predictions on one test set (i.e., no need to save the model object for future use), then the brif.trainpredict should be used.

Value

a data frame or a vector containing the prediction results. For regression, a numeric vector of predicted values will be returned. For classification, if type = "class", a character vector of the predicted class labels will be returned; if type = "score", a data frame will be returned, in which each column contains the probability of the new case being in the corresponding class.

Examples

# Predict using a model built by brif
pred_score <- predict(brif(Species ~ ., data = iris), iris, type = 'score')
pred_label <- predict(brif(Species ~ ., data = iris), iris, type = 'class')

# Equivalently and more efficiently:
pred_score <- brif(Species ~., data = iris, newdata = iris, type = 'score')
pred_label <- brif(Species ~., data = iris, newdata = iris, type = 'class')

# Or, retrieve predicted labels from the scores:
pred_label <- colnames(pred_score)[apply(pred_score, 1, which.max)]