confuMat-methods: Compute the confusion matrix for a classifier.

In lgatto/MLInterfaces: Uniform interfaces to R machine learning procedures for data in Bioconductor containers

Description

This function will compute the confusion matrix for a classifier's output

Methods

obj = "classifOutput", ...

Typically, an instance of class "classifierOutput" is built on a training subset of the input data. The model is then used to predict the class of samples in the test set. When the true class labels for the test set are available the confusion matrix is the cross-tabulation of the true labels of the test set against the predictions from the classifier. An optional t score threshold can also be specified.

obj = "classifierOutput", type="character", ...

For instances of classifierOutput, it is possible to specify the type of confusion matrix desired. The default is test, which tabulates classes from the test set against the associated predictions. If type is train, the training class vector is tabulated against the predictions on the training set. An optional t score threshold can also be specified.

obj = "classifierOutput", type="numeric"

For instances of classifierOutput, it is possible to specify the minimum score feature classification threshold. Features with a score less than the threshold are classified as NA in the confustion train or test confusion matrix.

Examples

library(golubEsets)
data(Golub_Merge)
smallG <- Golub_Merge[101:150,]
k1 <- MLearn(ALL.AML~., smallG, knnI(k=1), 1:30)
confuMat(k1)
confuMat(k1, "train")

lgatto/MLInterfaces documentation built on May 21, 2019, 5:12 a.m.