CS 445 Machine Learning
Fall 2020
Week 2 Study Guide
Learning Objectives
Decision Trees
- Define and compute the impurity of a set using entropy (IDD 3.3.3)
- Define and compute information gain (Δ info) (IDD 3.3.3)
- Utilize Δ info and manually create a decision tree using Hunt's algorithm (see IDD 3.3.1 and 3.3.4)
- Define a decision tree algorithm for classification
- Define a decision tree algorithm for regression
- Utilizing Seaborn and matplotlib to visual feature distributions
- Interpret a boxplot
- Define/compute quartiles on a set of data using Python
- Utilize Scikit-Learn to build and visualize a decision tree
Labs
- Decision Tree Intro
- Data Analysis and SkLearn Decision Trees
Resources
- IDD Sections 2.1, 3.3
- Scikit Learn Decision Tree Documentation
- Lecture slides
- BoxPlot Tutorial from Towards Data Science
Deliverables
| Topic | Description |
| Reading Quiz | Complete the reading quiz 1 on Canvas by the due date (indicated in Canvas) |
| PA 0 | Complete programming assignment 0 and submit the appropriate portions to Autolab and Canvas by the due date. |
In Progress
| Topic | Description |
| PA 1 | Decision Tree programming assignment. See the due date in Canvas. |