ddalphaf.getErrorRateCV: Test Functional DD-Classifier

In ddalpha: Depth-Based Classification and Calculation of Data Depth

Test Functional DD-Classifier

Description

Performs a cross-validation procedure over the given data. On each step every numchunks observation is removed from the data, the functional DD-classifier is trained on these data and tested on the removed observations.

Usage

ddalphaf.getErrorRateCV (dataf, labels, numchunks = 10, disc.type = c("LS", "comp"),  ...)

Arguments

dataf	list containing lists (functions) of two vectors of equal length, named "args" and "vals": arguments sorted in ascending order and corresponding them values respectively
labels	list of output labels of the functional observations
numchunks	number of subsets of testing data. Equals to the number of times the classifier is trained.
disc.type	type of the used discretization scheme. "LS" for ddalphaf.train, "comp" for for compclassf.train
...	additional parameters passed to ddalphaf.train

Value

errors	the part of incorrectly classified data
time	the mean training time
time_sd	the standard deviation of training time

See Also

ddalphaf.train to train the functional DD\alpha-classifier, ddalphaf.classify for classification using functional DD\alpha-classifier, ddalphaf.test to test the functional DD-classifier on particular learning and testing data, ddalphaf.getErrorRatePart to perform a benchmark study of the functional DD-classifier on particular data.

Examples

# load the fdata
df = dataf.growth()

stat <- ddalphaf.getErrorRateCV(dataf = df$dataf, labels = df$labels, 
                                numchunks = 5,
                                adc.args = list(instance = "avr", 
                                                numFcn = 2, 
                                                numDer = 2))
                                                
cat("Classification error rate: ", stat$errors, ".\n", sep = "")

    

ddalpha documentation built on Oct. 1, 2024, 1:07 a.m.