LDLcalc (version 2.1)

LDL_ML_Main_StackingAlgorithm: Create, train, assess and return a Stacking Algorithm Machine Learning prediction model

Description

This function reads data from a DATACSV.csv or data table file. It partitions them according to the partition parameter and labels them, trains all of the models and 'stacks' them into one, assesses them using the second set of the partition data, optionally plots some info relating the accuracy of the models and returns them for further use.

Usage

LDL_ML_Main_StackingAlgorithm(
  DataCSV,
  partition,
  ReportMultiPlot = TRUE,
  ComparisonPlot = TRUE
)

Value

It initializes and returns the stacked algorithm prediction model. In case of bad input, it will return -2 (illegitimate partition input)

Arguments

DataCSV

The .csv or data table file, path containing the data with which the model will be trained and assessed. Must contain at least 4 columns, named "CHOL", "HDL", "TG" and "LDLd", through which the train data and the validation data will be extracted.

partition

A value in the range (0,1) that stipulates what percentage of the input data will be used for training the model, while the remainder will be used to assess it.

ReportMultiPlot

A boolean that allows the user to select whether the LDL_ML_Main function will plot a diagram with 5 plots, relating different stats on the newly created model. Preset to TRUE.

ComparisonPlot

A boolean that allows the user to select whether the LDL_ML_Main_All_Models function will plot a comparison plot, relating different stats on the newly created models. Preset to TRUE.

Examples

# \donttest{
stackModel = LDL_ML_Main_StackingAlgorithm(SampleData,0.8,ReportMultiPlot=TRUE,ComparisonPlot=TRUE)
# }