Phase 1: Basics

Learn how to load built-in datasets from sklearn.datasets. These datasets return numpy arrays, which we'll convert to Polars DataFrames for easier exploration.

Use Polars DataFrames for fast, expressive data exploration and analysis before modeling.

Learn to split data into training and test sets to evaluate model performance on unseen data.

Master sklearn's core API pattern: create a model, fit it on training data, then predict on new data.

Learn K-Nearest Neighbors classification - a simple but powerful algorithm that classifies based on similarity to nearby points.

Learn Decision Trees - interpretable models that make predictions by learning simple decision rules from features.

Learn to evaluate classifiers using accuracy - the proportion of correct predictions.

Understand confusion matrices - a detailed breakdown of correct and incorrect predictions for each class.

Learn linear regression for predicting continuous values, using the same fit/predict pattern.

Learn to use trained models to make predictions on completely new, unseen data.

Learn how trained models can be saved and loaded for later use (conceptually - we'll use Python's pickle-like approach).

Understand how sklearn models handle multiple features and how to inspect feature importance.

Understand the difference between categorical and numerical features and why preprocessing matters.

Introduction to data preprocessing: scaling features to improve model performance.

Apply everything you've learned: load data, preprocess, train, and evaluate a complete ML pipeline.