DS Unit 2 Sprint 5: Linear Models

Welcome to Linear Models!

In this sprint, you'll learn how to build and evaluate linear models for regression and classification. You'll understand the mathematical foundations of these models, implement them using scikit-learn, and learn how to interpret their results.

Linear models form the foundation of many machine learning techniques. Despite their simplicity, they're powerful tools that can provide interpretable results and serve as baselines for more complex models.

Modules

This sprint is structured to provide you with a comprehensive understanding of linear models:

Module 1

Linear Regression 1

Learn the fundamentals of linear regression, starting with simple baselines. You'll implement linear regression using scikit-learn and understand how to interpret model coefficients.

View Module

Module 2

Linear Regression 2

Dive deeper into linear regression, learning about train-test splits, multiple regression, ordinary least squares, and the bias-variance tradeoff.

View Module

Module 3

Ridge Regression

Build on your regression knowledge with ridge regression. You'll learn about one-hot encoding, feature selection, and how regularization can improve model performance.

View Module

Module 4

Logistic Regression

Transition from regression to classification with logistic regression. You'll implement train-validate-test splits, understand classification baselines, and learn about scikit-learn pipelines.

View Module

Sprint Resources

Primary Resources

Code-Along Sessions Sprint Challenge

Documentation

Scikit-learn: Linear Models Scikit-learn: Linear Regression Scikit-learn: Ridge Regression Scikit-learn: Logistic Regression

GitHub Repositories

DS Unit 2 Linear Models DS Code-Alongs Unit 2

How to Succeed in This Sprint

  1. Establish baselines: Always start with simple baselines before building more complex models.
  2. Visualize your data: Understanding the relationships in your data is crucial for building effective models.
  3. Properly evaluate models: Use train-test splits and appropriate metrics to assess model performance.
  4. Interpret your results: Don't just build models—understand what the coefficients and predictions mean.
  5. Practice regularly: The more you work with these models, the better you'll understand their nuances.