jqm_cvi's Introduction

jqmcvi

Small module with Cluster Validity Indices (CVI)

Dunn and Davius Bouldin indices are implemented. It follows the equations presented in theory.pdf.

base.py : Python + NumPy

basec.pyx : Python + NumPy optimized with Cython

basec.pyx tested on M1 Macbook Pro, Python 3.9 and compiled with python setup.py build_ext -i

Functions:

dunn(k_list):

Slow implementation of Dunn index that depends on numpy

-- basec.pyx Cython implementation is much faster but slower than dunn_fast()

	""" Dunn index [CVI]
  
    Parameters
    ----------
    k_list : list of np.arrays
        A list containing a numpy array for each cluster |c| = number of clusters
        c[K] is np.array([N, p]) (N : number of samples in cluster K, p : sample dimension)
    """

dunn_fast(points, labels):

Fast implementation of Dunn index that depends on numpy and sklearn.pairwise

-- No Cython implementation

	""" Dunn index - FAST (using sklearn pairwise euclidean_distance function)
  
    Parameters
    ----------
    points : np.array
        np.array([N, p]) of all points
    labels: np.array
        np.array([N]) labels of all points
    """

davisbouldin(k_list, k_centers):

Implementation of Davis Boulding index that depends on numpy

-- basec.pyx Cython implementation is much faster

	""" Davis Bouldin Index

	Parameters
    ----------
    k_list : list of np.arrays
        A list containing a numpy array for each cluster |c| = number of clusters
        c[K] is np.array([N, p]) (N : number of samples in cluster K, p : sample dimension)
    k_centers : np.array
        The array of the cluster centers (prototypes) of type np.array([K, p])
    """

Developing

Install dependencies
- pip install -r requirements.txt
Recompile basec.c based on updated dependencies, updated Python version, or changes to basec.pyx
- python setup.py build_ext -i
(Re)Install jqmcvi after any changes
- pip install -e .