Silhouette Score Python Packages

manhattan-silhouette

Fast Silhouette computation under Manhattan Distance for clustering

optimalcluster

OptimalCluster is the Python implementation of various algorithms to find the optimal number of clusters. The algorithms include elbow, elbow-k_factor, silhouette, gap statistics, gap statistics with standard error, and gap statistics without log. Various types of visualizations are also supported.

cake-ensemble

Confidence in Assignments via K-partition Ensembles

sil-score

Exact and approximate silhouette scoring with micro, macro, and weighted cluster averages.

composite-silhouette

A clustering evaluation framework that combines micro- and macro-averaged silhouette scores into a composite metric using statistical weighting.

compsil

CompSil: Composite Silhouette for Cluster-Count Selection

silhouette-upper-bound

An upper bound of the Average Silhouette Width

clusopt-core

Clustream, Streamkm++ and metrics utilities C/C++ bindings for python