Module overview

  1. Understand how some selected machine learning models work: linear regression, decision trees, random forests, and gradient boosting.
  2. Look at how they are constructed and trained.
  3. Learn about gradient descent, an optimization algorithm widely used to train machine learning models.
  4. Become familiar with learning curves, bias / variance tradeoff, regularization, decision boundaries, and ensemble of models.
  5. Learn different ways to optimize hyperparameters.
  6. Get started with ML assignment 2

Reading

  • Chapter 4: Training Models
  • Chapter 6: Decision Trees
  • Chapter 7: Ensemble Learning and Random Forests

TODOs

Week 38

  1. Work with chapter 3: Classification.
  2. Work with the notebook DAT158-Part1-3-classification.ipynb.
  3. Work with assignment 1.
  4. Superficial reading of chapter 4.
  5. Look at the notebook DAT158-Part2-1-LinReg_GradientDescent.ipynb, DAT158-Part2-2-PolyReg_learning_curves_bias-variance.ipynb and DAT158-Part2-3-Regularization.ipynb.

Week 39

  1. Go through chapter 4: Training Models.
  2. Work on the notebooks: DAT158-Part2-1-LinReg_GradientDescent.ipynb, DAT158-Part2-2-PolyReg_learning_curves_bias-variance.ipynb and DAT158-Part2-3-Regularization.ipynb. Important to work on the "Your turn!" Tasks.

Week 40

  1. Presentation of assignment 1
  2. Go through chapter 6 and 7.
  3. Work with the notebooks: DAT158-Part2-4-Decision_Trees.ipynb and DAT158-Part2-5-Ensembling-RandomForests-and-Boosting.ipynb
  4. Extra material: DAT158-Part2-6-Extra-RandomForests-examples.ipynb and DAT158-Part2-6-Extra-Hyperparameter_optimization.ipynb
  5. Get started with assignment 2.

Lecture slides

Lecture videos

Notebook videos

Walkthrough DAT158-Part2-1-LinReg_GradientDescent.ipynb

Walkthrough DAT158-PyCaret-regression-example.ipynb