Machine Learning for iOS Developers

Chapter 8Creating a Decision Tree Classifier

WHAT'S IN THIS CHAPTER

• Training a decision tree classifier with Scikit-learn
• Examining the decision tree generated by the classifier
• Using the decision tree to find out the relative importance of each feature in the dataset
• Using Core ML tools to convert a Scikit-learn model into the Core ML format
• Using the decision tree model in an iOS application

In this chapter, you will create a decision tree classification model that can be used to classify Iris flowers. The model will be trained on the popular Iris flowers dataset. After the model is trained, you will use Core ML tools to export the model into the Core ML file format and finally use the model in an iOS app that allows users to input the characteristics of an Iris flower and learn what species it is likely to represent.

NOTE  You can download the code files for this chapter from wrox.com or from GitHub using the following URL:

https://github.com/asmtechnology/iosmlbook-chapter8.git

Decision Tree Recap

Decision trees were introduced briefly in Chapter 1, and as their name suggests, they are tree-like structures where each node represents a decision boundary and where leaves represent outcomes of the decision. Tree-based algorithms such as decision trees, random forests, and boosted trees are commonly used with real-world classification problems and are often the winning algorithms behind several Kaggle.com competitions based on tabular data.

The following are some ...