Feature engineering

Watch, read the card, then practise. Scaling · Encoding categories · Choosing features

Video coming soon

The video for this lesson is being recorded.

Meanwhile, the notes below and the 60-second cards cover the same ideas.

Scaling
Scaling puts features on similar ranges, so no feature dominates just because of its units.
Remember: Fit the scaler on training data only, then apply it to test data.
Encoding categories
Models need numbers, so text categories must be encoded.
Remember: No natural order → one-hot. Real order → ordinal.
Choosing features
Feature selection keeps the inputs that help and drops the ones that add noise or leak the answer.
Remember: Ask: would I know this feature at prediction time?

Try the example in Code Lab

import numpy as np
x = np.array([10, 30, 60])
print(((x - x.min()) / (x.max() - x.min())).round(2).tolist())

Lesson 2 of 4

Watch, read the card, then practise. Scaling · Encoding categories · Choosing features

Video coming soon

The video for this lesson is being recorded.

Meanwhile, the notes below and the 60-second cards cover the same ideas.

Scaling
Scaling puts features on similar ranges, so no feature dominates just because of its units.
Remember: Fit the scaler on training data only, then apply it to test data.
Encoding categories
Models need numbers, so text categories must be encoded.
Remember: No natural order → one-hot. Real order → ordinal.
Choosing features
Feature selection keeps the inputs that help and drops the ones that add noise or leak the answer.
Remember: Ask: would I know this feature at prediction time?

import numpy as np
x = np.array([10, 30, 60])
print(((x - x.min()) / (x.max() - x.min())).round(2).tolist())