Model evaluation and selection

We will begin by creating our training and testing sets, then create a random forest classifier as our base model. After evaluating its performance, we will move on and try the one-versus-rest classification method and see how it performs. We split our data 70/30. Also, one of the unique things about the mlr package is its requirement to put your training data into a "task" structure, specifically a classification task. Optionally, you can place your test set in a task as well.

A full list of models is available here, plus you can also utilize your own:

https://mlr-org.github.io/mlr-tutorial/release/html/integrated_learners/index.html

    > library(caret) #if not already loaded

    > set.seed(502)

    > split <- createDataPartition(y = df$class, p = 0.7, list = F)

    > train <- df[split, ]

    > test <- df[-split, ]

    > wine.task <- makeClassifTask...