Overview and warm-up

Day 2 uses yesterday’s object and array reasoning to organise behavioural data, write analysis functions, and prepare inputs for models.

Today’s notebooks

• Notebook 1: Lexical decision data — open or download

• Notebook 3: Model workflow — open or download

• Notebook 4: NLP text features — open or download

Retrieval warm-up

Exercise 16 (Retrieve yesterday’s axis reasoning)

For an array shaped participants × trials × channels × time:

1. Which axis produces a participant-level average across trials?

2. What shape remains?

3. Which expression extracts channel 4 for all participants, trials, and times?

Hint

An aggregation removes the axis it operates on. A colon retains every value along an axis; an integer selects one position and removes that axis.

Solution to Exercise 16 (Retrieve yesterday's axis reasoning)

1. axis=1

2. participants × channels × time

3. data[:, :, 4, :]

Today’s pipeline

files
  → DataFrame / arrays
  → validation
  → clean transformations
  → participant-level summaries
  → feature matrix X and target y
  → model
  → evaluation

You will not learn every model in one workshop. The useful common ground is the Python structure underneath regression, classification, NLP, and cognitive modelling.