RaSEn (version 3.0.0)

predict.super_RaSE: Predict the outcome of new observations based on the estimated super RaSE classifier (Zhu, J. and Feng, Y., 2021).

Description

Predict the outcome of new observations based on the estimated super RaSE classifier (Zhu, J. and Feng, Y., 2021).

Usage

# S3 method for super_RaSE
predict(object, newx, type = c("vote", "prob", "raw-vote", "raw-prob"), ...)

Arguments

object

fitted 'super_RaSE' object using Rase.

newx

a set of new observations. Each row of newx is a new observation.

type

the type of prediction output. Can be 'vote', 'prob', 'raw-vote' or 'raw-prob'. Default = 'vote'.

• vote: output the predicted class (by voting and cut-off) of new observations. Avalilable for all base learner types.

• prob: output the predicted probabilities (posterior probability of each observation to be class 1) of new observations. It is the average probability over all base learners.

• raw-vote: output the predicted class of new observations for all base learners. It is a n by B1 matrix. n is the test sample size and B1 is the number of base learners used in RaSE. Avalilable for all base learner types.

• raw-prob: output the predicted probabilities (posterior probability of each observation to be class 1) of new observations for all base learners. It is a n by B1 matrix.

...

additional arguments.

Value

depends on the parameter type. See the list above.

References

Zhu, J. and Feng, Y., 2021. Super RaSE: Super Random Subspace Ensemble Classification. https://www.preprints.org/manuscript/202110.0042

See Also

Rase.

Examples

# NOT RUN {
set.seed(0, kind = "L'Ecuyer-CMRG")
train.data <- RaModel("classification", 1, n = 100, p = 50)
test.data <- RaModel("classification", 1, n = 100, p = 50)
xtrain <- train.data$x
ytrain <- train.data$y
xtest <- test.data$x
ytest <- test.data$y

# fit a super RaSE classifier by sampling base learner from kNN, LDA and
# logistic regression in equal probability
fit <- Rase(xtrain = xtrain, ytrain = ytrain, B1 = 100, B2 = 100,
base = c("knn", "lda", "logistic"), super = list(type = "separate", base.update = T),
criterion = "cv", cv = 5, iteration = 1, cores = 2)
ypred <- predict(fit, xtest)
mean(ypred != ytest)
# }
# NOT RUN {
# }