← All notes

Answer: An estimator is an object that fits a model based on some training data and is capable of inferring some properties on new data. Scikit-learn provides dozens of built-in machine learning algorithms and models, called estimators. (S1, S2)

Using estimators

• Each estimator can be fitted to some data using its fit method. (S2)
• Both X and y are usually expected to be numpy arrays or equivalent array-like data types, though some estimators work with other formats such as sparse matrices. (S2)
• Once the estimator is fitted, it can be used for predicting target values of new data. (S2)

Writing your own

• Out of all the methods that an estimator implements, fit is usually the one you want to implement yourself. (S1)
• Common methods are implemented in BaseEstimator, which you should inherit from. (S1)
• A user should be able to instantiate an estimator without passing any arguments to it. (S1)
• Most arguments correspond to hyperparameters describing the model or the optimisation problem the estimator tries to solve. (S1)
• Meta-estimators are estimators that wrap other estimators. (S1)

Claims

• id: C1 claim: An estimator fits a model to training data. source: S1 evidence: "object that fits a model based on some training data and is capable of"
• id: C2 claim: Inherit from BaseEstimator. source: S1 evidence: "are implemented in BaseEstimator, which you should inherit from."
• id: C3 claim: Estimators are fitted with fit. source: S2 evidence: "Each estimator can be fitted to some data"
• id: C4 claim: Fitted estimators predict on new data. source: S2 evidence: "it can be used for predicting target values of"

Sources

• [S1] Developing scikit-learn estimators — scikit-learn 1.9.1 documentation. scikit-learn docs. https://scikit-learn.org/stable/developers/develop.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] Getting Started — scikit-learn 1.9.1 documentation. scikit-learn docs. https://scikit-learn.org/stable/getting_started.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_jkmykwwq,
  author = {{iinventedlight}},
  title = {{The scikit-learn estimator API}},
  howpublished = {iinventedlight Library},
  year = {2026},
  date = {2026-11-04},
  url = {https://iinventedlight.com/library/notes/sklearn-estimator-api/},
  note = {Permanent ID iil:note:jkmykwwq. 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-estimator-api.md.