balanced.train.and.test: Get a balanced test and train dataset

In loose.rock: Tools for Survival Analysis and Data Science

Description

Get a balanced test and train dataset

Usage

balanced.train.and.test(..., train.perc = 0.9, join.all = TRUE)

Arguments

...	vectors of index (could be numeric or logical)
train.perc	percentage of dataset to be training set
join.all	join all index in the end in two vectors (train and test vectors)

Value

train and test index vectors (two lists if 'join.all = FALSE', two vectors otherwise)

Examples

set1 <- seq(20)
balanced.train.and.test(set1, train.perc = .9)
####
set.seed(1985)
set1 <- rbinom(20, prob = 3/20, size = 1) == 1
balanced.train.and.test(set1, train.perc = .9)
####
set1 <- c(TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,FALSE,
TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,TRUE,FALSE,TRUE)
set2 <- !set1
balanced.train.and.test(set1, set2, train.perc = .9)

Example output

$train
 [1]  1  2  3  4  5  6  7  9 10 11 12 13 14 15 16 17 18 19

$test
[1]  8 20

$train
[1]  9 20

$test
[1] 15

$train
 [1]  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 18 20

$test
[1]  1 17 19

loose.rock documentation built on April 30, 2021, 1:06 a.m.