Search⌘ K
Home
Courses
Data Science with R: Decision Trees and Random Forests

Performing Cross-Validation

Explore the process of performing cross-validation with tidymodels in R. Understand how to set up data folds, track model accuracy metrics like sensitivity and specificity, and interpret the bias-variance tradeoff to improve decision tree model performance.

We'll cover the following...

Coding the workflow

One of the many benefits of using the tidymodels family of R packages is the standardized approach it provides for coding machine learning workflows. The following code sets up a machine learning workflow for a CART classification decision tree:

  1. Preparing the training data.

  2. Declaring how the data should be used to train the model.

  3. Specifying the machine learning algorithm to be used.

  4. Orchestrating the workflow.

R
#================================================================================================
# Load libraries - suppress messages
#
suppressMessages(library(tidyverse))
suppressMessages(library(tidymodels))
suppressMessages(library(rattle))
#================================================================================================
# Load the Titanic training data and transform Embarked to a factor
#
titanic_train <- read_csv("titanic_train.csv", show_col_types = FALSE) %>%
  mutate(Sex = factor(Sex),
         Embarked = factor(case_when(
           Embarked == "C" ~ "Cherbourg",
           Embarked == "Q" ~ "Queenstown",
           Embarked == "S" ~ "Southampton",
           is.na(Embarked) ~ "missing")))
#================================================================================================
# Craft the recipe - recipes package
#
titanic_recipe <- recipe(Survived ~ Sex + Pclass + SibSp + Parch + Fare + Embarked, data = titanic_train) %>%
  step_num2factor(Survived,
                  transform = function(x) x + 1,
                  levels = c("perished", "survived")) %>%
  step_num2factor(Pclass,
                  levels = c("first", "second", "third"))
#================================================================================================
# Specify the algorithm - parsnip package
#
titanic_model <- decision_tree() %>%
  set_engine("rpart") %>%
  set_mode("classification")
#================================================================================================
# Set up workflow - workflow package
#
titanic_workflow <- workflow() %>%
  add_recipe(titanic_recipe) %>%
  add_model(titanic_model)

Setting up cross-validation

The vfold_cv() from the rsample package creates the folds to be used in cross-validation. The following code uses these parameters:

...