KMeans plot

gistfile1.py

import os
import sys
import csv

sys.path.insert(0, os.path.abspath('../miner/'))

import miner.utils
import miner.clustering

def render_to_file(space, clusters):
    
    colors = {0: 'r', 1: 'b', 2: 'g', 3: 'y', 4: 'b'}
    
    import numpy as np
    import matplotlib.pyplot as plt
    
    fig = plt.figure(1, figsize=(5, 5))
    
    ax = fig.add_subplot(111)
    ax.set_aspect(1)
    
    # Redo our clusters in NumPy
    for cluster in range(len(clusters)):
        color = colors[cluster]
        
        x = []
        y = []
        for point in clusters[cluster]['points']:               
            x.append(space.points[point][0])
            y.append(space.points[point][1])
    
        np.array(x)
        np.array(y)
        ax.scatter(x, y, color=color)
    
    ax.set_xlim(-2, 2)
    ax.set_ylim(-1, 3.5)

    plt.draw()
    plt.savefig('kmeans.png')
    plt.close()



os.system('rm kmeans*.png')

space = miner.utils.Space();
file = csv.reader(open('dataforkmeans.csv'))

for parts in file:
    space.point(parts[0], parts[1])

kmeans = miner.clustering.KMeans(3, space)
kmeans.converge(render=render_to_file)