Scale
Scaling puts numeric features on a chosen scale. Some estimators depend strongly on feature magnitudes, while tree splits often do not need the same treatment.
Understand the idea
StandardScaler learns means and scales from fit data. Transform later rows using those learned values instead of fitting again on future or test data. A pipeline keeps preprocessing tied to the model during cross-validation. Scaling is not a universal improvement; choose it from the estimator and feature meanings.
Where to try it
Run locally with scikit-learn and NumPy installed in your virtual environment (py -m pip install scikit-learn numpy). These examples are executed during course review, but the online runner does not currently preload scikit-learn. The small data is for learning the API, not evidence of production model quality.
A worked example
from sklearn.preprocessing import StandardScaler
scaler = StandardScaler().fit([[10], [20], [30]])
print(scaler.transform([[10], [20], [30]]).round(3).ravel().tolist())
Read the result
The transformed values are approximately [-1.225, 0.0, 1.225].
Try one change
Transform a new value without refitting, then explain why its value need not fall between -1 and 1.
For package options, see the scikit-learn documentation.