Feature scaling

q.transform.scale provides sklearn-compatible numeric scaling transformers. Use them inside q.transform.Pipeline so learned parameters are fitted only on fit partitions:

pipeline = q.transform.Pipeline(
    [("scale", q.transform.scale.StandardScaler())]
)
scaled = pipeline.fit_transform(dataset)

StandardScaler is provided by scikit-learn and exposed through the QRT namespace for a consistent public API.