Classical machine learning by doing — the estimator API, regression and regularization, the main classifiers, honest evaluation with the right metrics, cross-validation and hyperparameter tuning, and unsupervised learning. Every lesson is runnable Python.
Before you start
You will run real Python with scikit-learn (the lessons use its built-in example datasets, so there is nothing to download). No GPU or API keys required. Basic pandas from the Data Analysis course helps.
The scikit-learn API
scikit-learn’s power is one consistent interface: every model has fit and predict. Learn it once and you can train, use, and swap any algorithm in a single line.
Regression
Predict a continuous number — a price, a temperature, a demand. Fit linear and polynomial regression, read the coefficients, and measure error the right way.
Regularization: Ridge, Lasso & ElasticNet
Unregularized models overfit — they memorize noise. Ridge, Lasso, and ElasticNet penalize large coefficients to generalize better, and Lasso even selects features.
Classification Algorithms
Predict a category — spam or not, which digit, which disease. Meet the workhorse classifiers, from logistic regression to random forests and gradient boosting.
Evaluating Classifiers
Accuracy lies on imbalanced data. Use precision, recall, F1, the confusion matrix, and ROC-AUC to know what your classifier is really doing — and where it fails.
Cross-Validation & Tuning
A single train/test split is noisy. Cross-validation gives a stable performance estimate, and grid or random search finds the hyperparameters that make a model shine.
Unsupervised Learning
Without labels, you find structure instead of predicting a target. Cluster similar rows with K-Means and compress features with PCA to discover what the data contains.