GPU accelerated k-means clustering toolbox for MATLAB

GPU k-means clustering is a programming tool for fast clustering of multidimensional data in the Euclidean space aimed to be used through MATLAB.
It is developed in CUDA targeting NVIDIA's CUDA enabled GPUs.

K-means is a basic clustering algorithm widely used for clustering large amounts of data. It has been used in numerous applications in areas such
as statistical machine learning, data mining, image processing and many more. However the computation has always been considered of high arithmetic
complexity and time consuming.

This tool intends to overcome this problem and provide MATLAB users with a simple yet powerful interface for clustering large amounts of data.

The performance benefit is achieved by full exploitation of the parallel manycore GPU architecture.

Current version V 1.0 compiles under Linux and requires the installation of CUDA (any version) and a MATLAB mex compiler (any version).

Download: gpu_kmeans_v1.0.tar.gz

gpu_accelerated_k-means_toolbox_for_matlab.txt · Last modified: 2012/10/06 00:16 by gpapamak
Recent changes RSS feed Creative Commons License Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki