Welcome to cvi’s documentation!
These pages serve as the official documentation for the cvi Python package, the Python implementation of the ClusterValidityIndices.jl Julia package.
Cluster Validity Indices (CVIs) tackle the problem of judging the performance of an unsupervised/clustering algorithm without the availability of truth or supervisory labels, resulting in metrics of under- or over-partitioning. Furthermore, Incremental CVIs (ICVIs) are variants of these ordinarily batch algorithms that enable an online and computationally tractable method of evaluating the performance of a clustering algorithm as it clusters while being numerically equivalent to their batch counterparts.
The purpose of this package is to provide a home for the development and use of these CVIs and ICVIs.