Implementation:Scikit learn Scikit learn AgglomerationTransform

Knowledge Sources	Scikit_learn Scikit-learn Docs
Domains	Machine Learning, Clustering
Last Updated	2026-02-08 15:00 GMT

Overview

Concrete tool for performing feature agglomeration via the transformer interface, provided by scikit-learn.

Description

The AgglomerationTransform class is a mixin that provides transform and inverse_transform methods for feature agglomeration. It pools features into clusters using a configurable pooling function (defaulting to np.mean) and maps data to and from the reduced feature space defined by cluster labels.

Usage

Use this mixin when building hierarchical clustering estimators that need to reduce feature dimensionality by grouping correlated features into clusters. It is used as a base class by FeatureAgglomeration.

Code Reference

Source Location

• Repository: scikit-learn
• File: sklearn/cluster/_feature_agglomeration.py

Signature

class AgglomerationTransform(TransformerMixin):
    def transform(self, X):
    def inverse_transform(self, X):

Import

from sklearn.cluster._feature_agglomeration import AgglomerationTransform

I/O Contract

Inputs

Name	Type	Required	Description
X	array-like of shape (n_samples, n_features)	Yes	Input data matrix to transform or inverse-transform

Outputs

Name	Type	Description
Y	ndarray of shape (n_samples, n_clusters)	The pooled values for each feature cluster (from transform)
X_original	ndarray of shape (n_samples, n_features)	Reconstructed values mapped back to original feature space (from inverse_transform)

Usage Examples

Basic Usage

from sklearn.cluster import FeatureAgglomeration
from sklearn.datasets import make_classification

X, y = make_classification(n_features=20, random_state=0)
agglo = FeatureAgglomeration(n_clusters=5)
X_reduced = agglo.fit_transform(X)
print(X_reduced.shape)  # (100, 5)

Related Pages

• Principle:Scikit_learn_Scikit_learn_Clustering