readme

改进的k均值算法

#
# About
#

This k-means clustering code is a mex implementation of the ICML2003 paper 

@misc{ elkan03using,
  author = "C. Elkan",
  title = "Using the triangle inequality to accelerate kMeans",
  text = "C. Elkan. Using the triangle inequality to accelerate kMeans. In Proceedings
    of the Twentieth International Conference on Machine Learning, 2003, pp.
    147-153.",
  year = "2003",
  url = "citeseer.ist.psu.edu/elkan03using.html" 
}


#
# Installation
#
You need to update the Makefile if you are on a 64bit machine (you will see it). and to give the correct
$MATLABDIR if you want to use matlab. Then call "make clean all shared"

Or within Matlab, using mex:
mex -cO mpi_kmeans.cxx
mex -O mpi_kmeans_mex.cxx mpi_kmeans.o
mex -O mpi_assign.cxx mpi_kmeans.o

If you are fine with single precision you can update the variable INPUT_TYPE in mpi_kmeans.h

#
# Use
#
Matlab:
Try "help mpi_kmeans" in a matlab shell. This will also give an example

Python: 
Just run ./mpi_kmeans.py for an example.


Tuebingen, 14th Feb 2008
Peter Gehler, [e-mail]