Time for action – clustering points

We will generate some random points and cluster them, which means that points that are close to each other are put in the same cluster. This is only one of the many techniques that you can apply with scikit-learn. Clustering is a type of machine learning algorithm, which aims to group items based on similarities. Second, we will calculate a square affinity matrix. An affinity matrix is a matrix containing affinity values; for instance, distances between points. Finally, we will cluster the points with the AffinityPropagation class from scikit-learn. Perform the following steps to cluster points:

  1. We will generate 30 random point positions within a square of 400 x 400 pixels:
    positions = np.random.randint(0, 400, size=(30, 2))
  2. We will use the Euclidean distance to the origin as affinity matrix.
    positions_norms = np.sum(positions ** 2, axis=1)
S = - positions_norms[:, np.newaxis] - positions_norms[np.newaxis, :] + 2 * np.dot(positions, positions.T)
  3. Give the AffinityPropagation class the result from the previous step. This class labels the points with the appropriate cluster number.
    aff_pro = sklearn.cluster.AffinityPropagation().fit(S)
labels = aff_pro.labels_
  4. We will draw polygons for each cluster. The function involved requires a list of points, a color (let's paint it red), and a surface.
    pygame.draw.polygon(screen, (255, 0, 0), polygon_points[i])

    The result is a bunch of polygons for each cluster, as shown in the following screenshot:

    Time for action – clustering points

    The clustering example code is as follows:

    import numpy as np
import sklearn.cluster
import pygame, sys
from pygame.locals import *

positions = np.random.randint(0, 400, size=(30, 2))

positions_norms = np.sum(positions ** 2, axis=1)
S = - positions_norms[:, np.newaxis] - positions_norms[np.newaxis, :] + 2 * np.dot(positions, positions.T)

aff_pro = sklearn.cluster.AffinityPropagation().fit(S)
labels = aff_pro.labels_

polygon_points = []

for i in xrange(max(labels) + 1):
    polygon_points.append([])

# Sorting points by cluster
for i in xrange(len(labels)):
    polygon_points[labels[i]].append(positions[i])

pygame.init()
screen = pygame.display.set_mode((400, 400))

while True: 
    for i in xrange(len(polygon_points)):
      pygame.draw.polygon(screen, (255, 0, 0), polygon_points[i])

    for event in pygame.event.get():
      if event.type == QUIT:
         pygame.quit()
         sys.exit()

    pygame.display.update()

What just happened?

The most important lines in the artificial intelligence example are described in more detail in the following table:

Function

Description

sklearn.cluster.AffinityPropagation().fit(S)

This creates an AffinityPropagation object and performs a fit using an affinity matrix.

pygame.draw.polygon(screen, (255, 0, 0), polygon_points[i])

This draws a polygon given a surface, a color (red in this case), and a list of points.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
3.143.3.208