standardize: Standardize training data and test data

In SwanCygnus/predata: What the Package Does (One Line, Title Case)

Description

Standardize training and test data by the information of training data. Test data is standardized using the information of training data.

Usage

standardize(test, train, y_method = c("center", "scale"))

Arguments

test	Test data to standardize.
train	Training data that is standardized and used to standardize test data.
y_method	Decide how outcome will be standardized. If FALSE, nothing is applied.

Details

Standardization will be done by following formula: z = (x - mean(x)) / sd(x)

Value

A list that has standardized training and test data, which includes standard deviation and mean of outcome, The outcome of test data is not changed.

train	Standardized training data
test	Standardized test data
y_mean	A value used for centering (mean of outcome in training data)
y_sd	A value used for scaling (standard deviation of outcome in test data)

Examples

data_train <- data.frame(y = rnorm(10), x1 = rnorm(10), x2 = rnorm(10))
data_test <- data.frame(y = rnorm(10), x1 = rnorm(10), x2 = rnorm(10))
apply(data_train, 2, mean); apply(data_train, 2, sd)
data_scaled <- standardize(test = data_test,
train = data_train, y_method = "scale")
apply(data_scaled$train, 2, mean); apply(data_scaled$train, 2, sd)

SwanCygnus/predata documentation built on July 1, 2020, 12:09 a.m.