Implementation:Scikit learn Scikit learn ConsensusScore

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

Overview

Concrete tool for computing the consensus score between two sets of biclusters, provided by scikit-learn.

Description

The consensus_score function computes the similarity between two sets of biclusters. It uses a pairwise similarity metric (default Jaccard coefficient) to compare biclusters, then applies the Hungarian algorithm (linear sum assignment) to find the optimal one-to-one matching between the two sets. The final score is the mean similarity over the best matching pairs.

Usage

Use this function to evaluate biclustering results by comparing predicted biclusters against ground truth or comparing two different biclustering solutions.

Code Reference

Source Location

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

Signature

@validate_params(
    {
        "a": [tuple],
        "b": [tuple],
        "similarity": [callable, StrOptions({"jaccard"})],
    },
    prefer_skip_nested_validation=True,
)
def consensus_score(a, b, *, similarity="jaccard"):

Import

from sklearn.metrics import consensus_score

I/O Contract

Inputs

Name	Type	Required	Description
a	tuple of (rows, columns)	Yes	First set of biclusters as (row_indicators, column_indicators)
b	tuple of (rows, columns)	Yes	Second set of biclusters as (row_indicators, column_indicators)
similarity	str or callable	No	Similarity metric to use (default 'jaccard')

Outputs

Name	Type	Description
score	float	Consensus score: mean similarity over the best matching bicluster pairs

Usage Examples

Basic Usage

import numpy as np
from sklearn.metrics import consensus_score

# Two sets of biclusters represented as (rows, columns)
a_rows = np.array([[True, True, False], [False, True, True]])
a_cols = np.array([[True, False], [False, True]])
b_rows = np.array([[True, True, False], [False, False, True]])
b_cols = np.array([[True, False], [False, True]])

score = consensus_score((a_rows, a_cols), (b_rows, b_cols))
print(score)

Related Pages

• Principle:Scikit_learn_Scikit_learn_Cluster_Evaluation