my_knn_cv: k-Nearest Neighbors Cross-Validation

In marquesjchacon/myfirstpackage: Package Building Demonstration

Description

This function predicts the output class from a given set of covariates and actual class values using the k-nearest neighbors approach of identifying patterns in data points near each other.

Usage

my_knn_cv(train, cl, k_nn, k_cv)

Arguments

train	An input data frame with all the values of the predictor variables.
cl	The (actual) response values corresponding to each of the values from train.
k_nn	A numeric integer representing the number of neighbors considered.
k_cv	A numeric integer representing the number of folds for which to divide the observations in train.

Value

A list with elements class and cv_err, where class is a vector of the predicted classes for all observations, and cv_err is a numeric with the cross-validation misclassification error.

Examples

my_knn_cv(my_iris[, 1:4], as.data.frame(my_iris[, 5]), 1, 5)
my_knn_cv(my_iris[, 1:4], as.data.frame(my_iris[, 5]), 5, 5)

marquesjchacon/myfirstpackage documentation built on April 2, 2020, 9:42 p.m.