ClusterValidityIndices.jl review
A focused Julia toolkit for scoring clustering results without ground-truth labels.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
ClusterValidityIndices.jl is an open-source Julia package for evaluating unsupervised clustering without ground-truth labels. It is aimed at Julia users who need to calculate cluster-validity criteria from feature matrices and integer cluster-label vectors, whether they are comparing completed clustering results or updating evaluations incrementally. Its role is specific: it addresses clustering evaluation, not a broad range of statistical-analysis tasks.
The package’s main strength is its set of indices and two evaluation modes. The documented metrics include Calinski-Harabasz, centroid-based Silhouette, Davies-Bouldin, Generalized Dunn’s Index 43 and 53, Partition Separation, WB, Xie-Beni, and Renyi representative Cross Information Potential. Batch and online evaluation are both supported, and get_cvi! is the public evaluation API. That gives Julia workflows options for comparing partitions and for incremental assessment without changing to a general-purpose statistics product. It is a better fit when these named criteria matter than when a team needs a wider analysis environment.
ClusterValidityIndices.jl is distributed as a Julia package, with installation through Pkg or JuliaHub, and is available for Linux, Windows, and macOS. It is free and open source under the MIT license, with a self-hosted deployment model; the published product information describes community and documentation support. The narrow input format and Julia-specific package form are meaningful constraints for users whose work does not already fit that ecosystem. Julia users focused on clustering criteria should consider it; people seeking broader statistical analysis or a hosted deployment should look elsewhere.
ClusterValidityIndices.jl pros and cons
- Where it wins
- Evaluates cluster validity in both batch and incremental forms
- Includes Calinski-Harabasz, Silhouette, Davies-Bouldin, and other indices
- Open-source MIT package installable through Pkg or JuliaHub
- Where it doesn't
- Limited to clustering evaluation rather than general statistical analysis
- Works with feature matrices and integer cluster-label vectors
- Julia package rather than a hosted service, with community and documentation support
ClusterValidityIndices.jl fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update