Jump to content

Connect SuperML | Leeroopedia MCP: Equip your AI agents with best practices, code verification, and debugging knowledge. Powered by Leeroo — building Organizational Superintelligence. Contact us at founders@leeroo.com.

Implementation:Online ml River Metrics FowlkesMallows

From Leeroopedia


Knowledge Sources
Domains Online_Learning, Evaluation_Metrics, Clustering
Last Updated 2026-02-08 16:00 GMT

Overview

Fowlkes-Mallows Index for measuring similarity between two clusterings or classifications.

Description

FowlkesMallows computes the geometric mean of precision (PPV) and recall (TPR) to measure similarity between two clusterings. The formula is FM = sqrt(PPV × TPR) = sqrt((TP/(TP+FP)) × (TP/(TP+FN))). Higher values indicate greater similarity between the clusterings, with 1.0 representing perfect agreement and 0.0 representing no agreement.

Usage

Use FowlkesMallows Index to evaluate clustering algorithms by comparing predicted clusters with ground truth labels, or to measure agreement between two different clustering solutions. It's particularly useful as an external clustering evaluation metric that considers both precision and recall aspects of cluster assignments.

Code Reference

Source Location

Signature

class FowlkesMallows(metrics.base.MultiClassMetric):
    def __init__(self, cm=None):
        pass

Import

from river import metrics

I/O Contract

Method Parameters Returns Description
update y_true, y_pred None Updates metric with true and predicted cluster labels
get - float Returns Fowlkes-Mallows Index (0.0 to 1.0)

Note: This metric does not support sample weights (works_with_weights returns False).

Usage Examples

from river import metrics

y_true = [0, 0, 0, 1, 1, 1]
y_pred = [0, 0, 1, 1, 2, 2]

metric = metrics.FowlkesMallows()

for yt, yp in zip(y_true, y_pred):
    metric.update(yt, yp)
    print(metric)
# FowlkesMallows: 0.00%
# FowlkesMallows: 100.00%
# FowlkesMallows: 57.74%
# FowlkesMallows: 40.82%
# FowlkesMallows: 35.36%
# FowlkesMallows: 47.14%

# The final score of 47.14% indicates moderate agreement
# between the true and predicted clusterings

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment