CS 445 Machine Learning
Fall 2020
Week 3 Study Guide
Learning Objectives
- Define model underfitting and overfitting (IDD 3.4)
- Define training, validationb, and testing sets including their roles in proper experimental design and model selection(IDD 3.5)
- Define K-fold cross validation and design model evaluations that incorporate this technique (IDD 3.6)
- Define and contrast the difference between a model parameter and a model hyperparameter (IDD 3.7)
- Represent data in vector/matrix form with NumPy and process this data using NumPy functions (including array masking) instead of regular for/while loops (Numpy Lab, links to scipy lectures, and intro to numpy video)
- Define an algorithm for the K-Nearest Neighbors Classifier (IDD 4.3)
- Define/compute MinMaxScaling and characterize why it is important in the context of classifiers that utilize distances like KNN (IDD 2.4.1, lecture slides, and KNN Lab)
Resources
- Data Mining Textbook: 3.3 - 3.8
- Lecture Slides
Labs
Deliverables
| Topic | Description |
| Reading Quiz | Complete the reading quiz on Canvas by the due date (indicated in Canvas) |
| PA 1 | Submit code to Autolab< and documents are due in Canvas (see the due date in Canvas). |