Machine Learning Foundations

Train and evaluate practical models with careful data splits, useful baselines, and a clear understanding of model limits.

Your learning path

Problem framing and preparation

• Define the prediction task and target
• Train, validation, and test data
• Preprocessing and leakage prevention
• Start with a useful baseline

Training and evaluation

• Linear and logistic models
• Decision trees and random forests
• Classification metrics and business cost
• Cross-validation and parameter search

Reliable model delivery

• Imbalanced data and decision thresholds
• Error analysis and model interpretation
• Persisting a pipeline and monitoring drift
• Churn project review and model card

Build something useful

Build a customer churn classifier with a preprocessing pipeline, a baseline comparison, and a model evaluation report.

What to expect

Twelve guided reading lessons, worked examples, practice worksheets, a subject knowledge check, and an applied project. Allow approximately 40 minutes per lesson, with additional time for your project. The first lesson is available as a free preview.