FSPEC 101 - Machine Learning

Description

ML 101 provides a comprehensive introduction to machine learning, covering key algorithms, data preprocessing techniques, and model optimization. Students will learn how to implement K-Nearest Neighbors (KNN), gradient descent, logistic regression, and multi-class classification while using TensorFlow.js for real-world applications such as image recognition and data analysis. The course emphasizes hands-on projects, including CSV data handling, feature normalization, and performance optimization.

Prerequisite

Duration

None8 weeks

Curriculum

Week 1

Day 1

  • Algorithm Overview
  • How K-Nearest Neighbor Works
  • Implementing KNN

Day 2

  • Finishing KNN Implementation
  • Testing the Algorithm
  • Interpreting Bad Results

Day 3

  • Test and Training Data
  • Randomizing Test Data
  • Generalizing KNN

Day 4

  • Gauging Accuracy
  • Printing a Report
  • Refactoring Accuracy Reporting

Day 5

  • Investigating Optimal K Values
  • Updating KNN for Multiple Features
  • Multi-Dimensional KNN