####################################################################################################################################################

#Copyright (c) 2013, Josiah Walker

#All rights reserved.

#

#Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

#

# Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

# Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or #other materials provided with the distribution.

#

#THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED #WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY #DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS #OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING #NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

####################################################################################################################################################

"""

GPU based K Means algorithm.

"""

import time

from numpy import array,zeros,amax,amin,sqrt,dot

import numpy

import random

from numpy.linalg import eig

import pycuda.autoinit

import pycuda.driver as drv

from pycuda.compiler import SourceModule

import math

from DataUtils import dataConfig,loadTable,loadSplitTable,loadMatrix,loadIntMatrix,saveTable

KernelLocation = "CudaKernels/KMEANS/"

# KMeans Algorithm --------------------------------------------

def KMeansConfig(dataTable, k, eps = 0.00001, srcDims=100000000000,gpuMemSize = 512, settings = {}):

return settings

def KMeans(dataTable, k, epsilon=0.00001, srcDims = 1000000000000000, iters=20, normData = False):

return finalSols[results[0][1]]

if __name__ == '__main__':

import getopt,sys

arg_values = ['dims=','if=','of=','k=','eps=','help','h']

optlist, args = getopt.getopt(sys.argv[1:], 'x', arg_values)

srcDims = 10000000000000

k = -1

eps = 0.00001

infile='swissroll.csv'

outfile='swissmeans.csv'

for o in optlist:

if o[0].strip('-') == 'dims':

srcDims = int(o[1])

for o in optlist:

if o[0].strip('-') == 'if':

infile = o[1]

for o in optlist:

if o[0].strip('-') == 'of':

outfile = o[1]

for o in optlist:

if o[0].strip('-') == 'k':

k = int(o[1])

for o in optlist:

if o[0].strip('-') == 'eps':

eps = float(o[1])

if k < 0:

k = len(open(infile).readlines())/10

for o in optlist:

if o[0].strip('-') == 'help' or o[1].strip('-') == 'h':

print "The following commands are available:"

print "\t--if=inputfile\tDefaults to swissroll.csv"

print "\t--of=outputfile\tDefaults to swissmeans.csv"

print "\t--k=k_nearest_neighbours\tDefaults to dataset_length/10"

print "\t--dims=input_dimensions\tDefaults to all in the input file"

print "\t--eps=termination_epsilon\tDefaults to 0.00001"

result = KMeans(infile,k,eps,srcDims)

saveTable(result,outfile)