← All notes

Answer: The sklearn.preprocessing package provides several common utility functions and transformer classes to change raw feature vectors into a representation that is more suitable for the downstream estimators. (S1)

Standardization

• In practice we often ignore the shape of the distribution and just transform the data to center it by removing the mean value of each feature, then scale it by dividing non-constant features by their standard deviation. (S1)
• StandardScaler implements the Transformer API to compute the mean and standard deviation on a training set so as to be able to later re-apply the same transformation on the testing set. (S1)
• An alternative standardization is scaling features to lie between a given minimum and maximum value, often between zero and one; this can be achieved using MinMaxScaler. (S1)
• If some outliers are present in the set, robust scalers or other transformers can be more appropriate. (S1)

Use them consistently

• If these data transforms are used when training a model, they also must be used on subsequent datasets, whether it's test data or data in a production system. (S2)
• Both train and test subsets should receive the same preprocessing transformation, but these transformations should only be learnt from the training data. (S2)

Claims

• id: C1 claim: sklearn.preprocessing provides transformer classes. source: S1 evidence: "The sklearn.preprocessing package provides several common"
• id: C2 claim: Robust scalers suit data with outliers. source: S1 evidence: "If some outliers are present in the set, robust scalers or other transformers can"
• id: C3 claim: Transforms used in training must be used later too. source: S2 evidence: "must be used on subsequent datasets"
• id: C4 claim: MinMaxScaler or MaxAbsScaler do range scaling. source: S1 evidence: "This can be achieved using MinMaxScaler or MaxAbsScaler"

Sources

• [S1] 8.3. Preprocessing data — scikit-learn 1.9.1 documentation. scikit-learn docs. https://scikit-learn.org/stable/modules/preprocessing.html . License: BSD-3-Clause (scikit-learn/scikit-learn COPYING; Copyright (c) 2007-2026 The scikit-learn developers; https://github.com/scikit-learn/scikit-learn/blob/main/COPYING). Accessed 2026-11-04.
• [S2] 12. Common pitfalls and recommended practices — scikit-learn 1.9.1 documentation. scikit-learn docs. https://scikit-learn.org/stable/common_pitfalls.html . License: BSD-3-Clause (scikit-learn/scikit-learn COPYING; Copyright (c) 2007-2026 The scikit-learn developers; https://github.com/scikit-learn/scikit-learn/blob/main/COPYING). Accessed 2026-11-04.

License: this note is CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Summary by iinventedlight (AI-assisted).

@misc{iil_note_1jfztq6y,
  author = {{iinventedlight}},
  title = {{Feature scaling with sklearn.preprocessing}},
  howpublished = {iinventedlight Library},
  year = {2026},
  date = {2026-11-04},
  url = {https://iinventedlight.com/library/notes/sklearn-preprocessing-scaling/},
  note = {Permanent ID iil:note:1jfztq6y. License: CC-BY-4.0.},
}

Markdown

Spotted a mistake? Use the MCP feedback tool or email hello@iinventedlight.com.

For language models and agents: start at /llms.txt. This page in Markdown: /library/notes/sklearn-preprocessing-scaling.md.