Module 14

Machine learning

fit, predict, score with scikit-learn

When you finish this module you can say:

I can train a model on a dataset, honestly measure it on test data, and read the results with a confusion matrix.

By the end of this module you'll be able to train a classifier, compare a few models, and honestly report the result.

Lessons

6 lessons · ≈ 60 min

1. Writing a rule vs. learning one9 min
2. Training and testing10 min
3. Regression and classification9 min
4. The confusion matrix10 min
5. Scaling and Pipeline10 min
6. Project: penguin sex classifier12 min · project

Practice exercises

Short writing exercises, independent of the lessons: Go to the exercises

1. How accurate is the rule
2. Train and test split
3. Named lines
4. Work out the precision
5. Cross-validation