| predict.rvbm {rpud} | R Documentation |
The predict.rvbm method predicts class
membership of given test data based on
the Gaussian process model created by rvbm.
## S3 method for class 'rvbm'
predict(object, X.TEST=NULL, ...)
object |
An S3 object that inherits from the class |
X.TEST |
A matrix of new test data. If NULL, the feature matrix in obj is used instead. |
... |
Not used. |
An vector containing the predicted class membership based on the maximum posterior probability.
An attribute named Ptest contains the predicted posterior probability.
This replaces the predictCPP and predClass methods in vbmp.
Chi Yau
chi.yau@r-tutor.com
rvbm, vbmp
## Not run:
library(rpud)
x <- rvbm.sample.train$X
y <- rvbm.sample.train$t.class
model.rvbm <- rvbm(
x, y, x, y,
theta = rep(1, ncol(x)),
control = list(
sKernelType="gaussian",
bThetaEstimate=TRUE,
bMonitor=TRUE,
InfoLevel=1)
)
res <- predict(model.rvbm, rvbm.test$X)
err <- sum(res != rvbm.test$t.class); err
rate <- err/length(rvbm.test$t.class)*100; rate
## End(Not run)