Development build. This page describes main, not a released package. The latest published Lodestar.Metrics is 0.3.0 — read its documentation.

Home › Metrics › Clustering metrics

FowlkesMallows

Agreement between two partitions as the geometric mean of pair precision and pair recall.

public static class FowlkesMallows

Example — the same partition under different names.

using Lodestar.Metrics;

double same = FowlkesMallows.Score([0, 0, 1, 1], [2, 2, 0, 0]);  // => 1

Remarks — counts pairs of samples, like AdjustedRand, and unlike it applies no correction for chance. Two independent partitions therefore score above zero here, which makes this the wrong measure for comparing clusterings with different numbers of clusters and a reasonable one for comparing two at the same size.

Reference behaviour is sklearn.metrics.fowlkes_mallows_score.

Applies to — net10.0, netstandard2.0.

See also — AdjustedRand, AdjustedMutualInformation, the clustering index.

Members

Member	What it does
FowlkesMallows.Score	The geometric mean of pair precision and pair recall.