Introduction

What is Buddi-CRAVeD ?

Buddi-CRAVeD (ClusteR Analysis and Validation for largE Datasets) is a easy-to-use Cluster Analysis, Visualisation and Validation suite. It draws upon the power of standard Python3 APIs for data analysis, machine learning and data visualisation.

The suite extends support for an extensive range of :

• Clustering Algorithms
• Internal and External Cluster Validation Indices
• Methods for visualisation of 2D and 3D Data

The suite along with our modified versions of the companion libraries (scikit-learn, scipy) allows us to push the limits on “maximum size of datasets for cluster analysis on common desktops and laptops” posed due low main memory availabilty. This feat is achieved by moving the memory-intensive data structures that cloge the main memory onto HDF5 files that can be put on hard-disks.

Who is it intended for ?

Buddi-CRAVeD is a “quick-starter” or “one-stop” for machine-learning enthusiasts seeking to explore the areas of Exploratory Data Analysis (EDA), Cluster Analysis and Cluster Validation with ease.

It also qualifies as a great companion tool to assist quality research on Cluster Validation and allied domains.

For those “craving” to do more (i.e., process larger datasets) with the limited main memory of their desktops and laptops, we help you push the boundaries.