multipleModels: Training multiple models for benchmark comparison

In nagdevAmruthnath/EnsembleML: An ensemble modeling package and deployment

Description

Training multiple models for benchmark comparison

Usage

multipleModels(train, test, y, metric, nfolds, repeats, models)

Arguments

train	A training data frame
test	A testing data frame
y	Response variable
metric	A minimization metric for training. If not mentioned, for regression RMSE and for classification Kappa value will be used.
nfolds	Number of kfolds for cross validation. By default, 5 will be used.
repeats	Number of repeats for cross validation. By default, 5 will be used.
models	A character list of models to train based on caret package structure

Value

summary of all trained models and trained models

Examples

#data("iris")
# multipleModels(train = iris, test = iris, y = "Species", models = c("C5.0", "parRF"))

## results
##      Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull AccuracyPValue McnemarPValue
##C5.0      1.00  1.00     0.9757074     1.0000000    0.3333333   2.702787e-72           NaN
##parRF     0.96  0.94     0.9149722     0.9851815    0.3333333   2.525127e-60           NaN

nagdevAmruthnath/EnsembleML documentation built on Nov. 5, 2019, 2:20 p.m.