Academy / Data & Machine Learning

Path 2 · Course 3 · For analysts and aspiring practitioners

Machine Learning Fundamentals

Build, evaluate and explain your first machine learning models with scikit-learn: framing problems, regression, classification and thresholds, overfitting and ensembles, cross-validation, clustering and PCA, and interpreting and deploying models responsibly.

📘 6 modules⏱ About 3.5 hours✅ Knowledge check per module🎓 Final assessment & certificate🎯 Python & pandas

0 of 6 modules complete

Start Module 1 Final assessment

Modules

  1. 1

    The machine learning workflow

    Framing problems, classification vs regression, the end-to-end workflow, scikit-learn's fit/predict pattern, the practice dataset and a first model against a baseline.

    30 min
  2. 2

    Regression: predicting numbers

    Linear regression, training and evaluating against a baseline, MAE, RMSE and R², interpreting coefficients, and inspecting errors.

    30 min
  3. 3

    Classification and evaluation

    Logistic regression and probabilities, decision trees, confusion matrices, precision, recall, F1 and ROC-AUC, choosing thresholds, and imbalanced classes.

    35 min
  4. 4

    Better models: overfitting, ensembles and validation

    Diagnosing overfitting, random forests and gradient boosting, k-fold cross-validation, hyperparameter tuning with grid search, and leakage-safe pipelines.

    35 min
  5. 5

    Unsupervised learning

    Segmentation with k-means, scaling, choosing the number of clusters, PCA for visualisation, and anomaly detection with isolation forests.

    30 min
  6. 6

    Interpreting and deploying models

    Permutation importance, error analysis by group, saving models safely, deployment and monitoring, model cards, and choosing between classic ML and LLMs.

    30 min