ddalpha (version 1.3.16)

ddalpha.getErrorRatePart: Test DD-Classifier

Description

Performs a benchmark procedure by partitioning the given data. On each of times steps size observations are removed from the data, the DD-classifier is trained on these data and tested on the removed observations.

Usage

ddalpha.getErrorRatePart(data, size = 0.3, times = 10,  ...)

Value

errors

the part of incorrectly classified data (mean)

errors_sd

the standard deviation of errors

errors_vec

vector of errors

time

the mean training time

time_sd

the standard deviation of training time

Arguments

data

Matrix containing training sample where each of $n$ rows is one object of the training sample where first $d$ entries are inputs and the last entry is output (class label).

size

the excluded sequences size. Either an integer between $1$ and $n$, or a fraction of data between $0$ and $1$.

times

the number of times the classifier is trained.

...

additional parameters passed to ddalpha.train

See Also

ddalpha.train to train the DD$\alpha$-classifier, ddalpha.classify for classification using DD$\alpha$-classifier, ddalpha.test to test the DD-classifier on particular learning and testing data, ddalpha.getErrorRateCV to get error rate of the DD-classifier on particular data.

Examples

# Generate a bivariate normal location-shift classification task
# containing 200 objects
class1 <- mvrnorm(100, c(0,0), 
                  matrix(c(1,1,1,4), nrow = 2, ncol = 2, byrow = TRUE))
class2 <- mvrnorm(100, c(2,2), 
                  matrix(c(1,1,1,4), nrow = 2, ncol = 2, byrow = TRUE))
propertyVars <- c(1:2)
classVar <- 3
data <- rbind(cbind(class1, rep(1, 100)), cbind(class2, rep(2, 100)))

# Train 1st DDalpha-classifier (default settings) 
# and get the classification error rate
stat <- ddalpha.getErrorRatePart(data, size = 10, times = 10)
cat("1. Classification error rate (defaults): ", 
    stat$error, ".\n", sep = "")

# Train 2nd DDalpha-classifier (zonoid depth, maximum Mahalanobis 
# depth classifier with defaults as outsider treatment) 
# and get the classification error rate
stat2 <- ddalpha.getErrorRatePart(data, depth = "zonoid", 
                                outsider.methods = "depth.Mahalanobis", size = 0.2, times = 10)
cat("2. Classification error rate (depth.Mahalanobis): ", 
    stat2$error, ".\n", sep = "")