Home
Courses
Machine Learning with Tree-Based Models in R

Visually inspect the tree model

Plot the tree and read it using fancyRpartPlot()

This exercise is part of the course

Machine Learning with Tree-Based Models in R

BeginnerSkill Level

0.0+

0 reviews

Start Course for Free

Supervised machine learning with classification trees.

Exercise 1: Welcome to the course!Exercise 2: Build a decision tree Exercise 3: Introduction to classification trees Exercise 4: Advantages of tree-based methods Exercise 5: Decision trees vs. linear models Exercise 6: Overview of the modeling process Exercise 7: Split the data Exercise 8: Train a classification tree model Exercise 9: Visually inspect the tree model

Current Exercise

Exercise 10: Evaluate model performance Exercise 11: What are the hyperparameters for a classification tree?Exercise 12: Tuning the model

Supervised machine learning with regression trees.

Exercise 1: Introduction to regression trees Exercise 2: Classification vs. regression Exercise 3: Split the data Exercise 4: Train a regression tree model Exercise 5: Performance metrics for regression Exercise 6: Evaluate regression tree model Exercise 7: Use of splitting criterion in trees Exercise 8: Compare models that use different criterion Exercise 9: Grid Search for model selection Exercise 10: Perform a Grid Search Exercise 11: Generate a grid of models Exercise 12: Evaluate the grid

In this chapter, we will present Bagged Trees, an ensemble method, that uses a combination of trees (instead of only one).

Exercise 1: Introduction to bagged trees Exercise 2: Advantages of bagged trees Exercise 3: Train a bagged tree model Exercise 4: Understanding the bagged model output Exercise 5: Out-of-bag (OOB) estimate of performance Exercise 6: Predict labels on a test set Exercise 7: Confusion matrix for test set Exercise 8: Using caret for cross-validating models Exercise 9: Cross-validate a bagged tree model in caret Exercise 10: Evaluate cross-validated model performance Exercise 11: Generate predictions from the caret model Exercise 12: Compare test set performance to CV performance

In this chapter, we will present the Random Forest algorithm, another ensemble method, which is a modified version of bagging (with decision trees). Here you'll learn how to train, tune and evaluate Random Forest models in R.

Exercise 1: Introduction to Random Forest Exercise 2: Bagged trees vs. Random Forest Exercise 3: Train a random forest model Exercise 4: Understanding random forest model output Exercise 5: Evaluate out-of-bag error Exercise 6: Evaluate model performance on a test set Exercise 7: OOB error vs. test set error Exercise 8: Advantages of OOB error Exercise 9: Evaluate test set AUC Exercise 10: Tuning a random forest model Exercise 11: Tuning a random forest via mtry Exercise 12: Tuning a random forest via tree depth

In this chapter, we will present the boosting methodology with a focus on the Gradient Boosting Machine (GBM) algorithm, another popular tree-based ensemble method. Here you'll learn how to train, tune and evaluate GBM models in R.

Exercise 1: Introduction to boosting Exercise 2: Bagged trees vs. boosted trees Exercise 3: Specify the distribution of the response variable Exercise 4: Train a GBM model Exercise 5: Understanding GBM model output Exercise 6: Prediction using a GBM model Exercise 7: Evaluate test set AUC Exercise 8: GBM hyperparameters Exercise 9: Early stopping in GBMs Exercise 10: Use caret to train a grid of GBMs Exercise 11: Select the best GBM from the grid Exercise 12: Bringing it all together