Notebooks

Open the repository in VS Code and select the workshop environment before running a notebook. Use Shift+Enter to run a cell.

Notebook 0: Python warm-up

Lists, dictionaries, comprehensions, functions, debugging, and short written answers.

• Open Notebook 0 in the book

• Download Notebook 0

Notebook 1: Lexical decision data

Analyse trial-level reaction times and accuracy from a lexical-decision experiment.

• Open Notebook 1 in the book

• Download Notebook 1

• Download the lexical-decision CSV

Notebook 2: EEG arrays

Move between pandas and NumPy, select samples with Boolean masks, calculate channel means, and practise reshaping arrays.

• Open Notebook 2 in the book

• Download Notebook 2

• Download the EEG Eye State data

Notebook 3: Model workflow

Create participant-level features from lexical-decision trials, split at the participant level, fit a scikit-learn pipeline, and inspect the confusion matrix and leakage risks.

• Open Notebook 3 in the book

• Download Notebook 3

Notebook 4: NLP text features

Build a document–term matrix and TF–IDF representation, compare documents, and transfer NumPy axis reasoning to token embeddings.

• Open Notebook 4 in the book

• Download Notebook 4

The checks compare selected values, shapes, or columns with expected results. Written interpretations are not checked automatically.

Solutions

Try each exercise notebook first. When you need to compare your approach, use the matching completed notebook:

• Notebook 0 solution

• Notebook 1 solution

• Notebook 2 solution

• Notebook 3 solution

• Notebook 4 solution

In the repository, all completed notebooks are grouped under notebooks/solutions/.

Data sources

The EEG Eye State data was donated by Oliver Roesler and is distributed by the UCI Machine Learning Repository under CC BY 4.0 (DOI).

The lexical-decision data is distributed with the languageR package and described in Baayen’s Analyzing Linguistic Data: A Practical Introduction to Statistics (2008). The package is licensed under GPL (≥ 2).