StatRed ass4: Implemented part 1 (with very nice dataset generator :)).

statred/ass3/classify.py

@@ -69,10 +69,10 @@ for i in range(4):
             facecolor=[1,1,1]*len(T))
         scatter(XC[T, i], XC[T, j], s=30, marker='+',
             edgecolor=color[c.astype(int) - 1])
-from pylab import show
-show()
-#if method == 'knnb':
-#    filename = 'knnb-%d.pdf' % k
-#else:
-#    filename = '%s.pdf' % method
-#savefig(filename)
+#from pylab import show
+#show()
+if method == 'knnb':
+    filename = 'knnb-%d.pdf' % k
+else:
+    filename = '%s.pdf' % method
+savefig(filename)

statred/ass4/k-means.py

+from pylab import loadtxt, array, scatter, figure, show, mean, argmin, \
+        ones, append, savefig
+from random import random, seed
+from sys import argv, exit
+
+def init(X, k):
+    return X[:k]
+
+def init_pp(X, k):
+    return X[:k]
+
+if len(argv) == 3:
+    if argv[2] != 'pp':
+        print 'Usage: python %s K [ "pp" ]' % argv[0]
+        exit()
+    print 'Using k-means++'
+    initial_means = init_pp
+else:
+    print 'Using normal k-means'
+    initial_means = init
+k = int(argv[1])
+
+# Generate dataset
+seed(700)
+n, N = 2, 100
+X = array([[100 * random() for j in range(n)] for i in range(int(N / k + N % k))])
+for c in range(k - 1):
+    d = (k + 1) * 100 * random()
+    X = append(X, [[100 * random() + d for j in range(n)] for i in \
+        range(int(N / k))], 0)
+
+M = initial_means(X, k)
+Mp = M - 1
+steps = 0
+
+# Divide in clusters
+while (Mp - M).any():
+    Mp = M
+    clusters = [[] for i in range(k)]
+    # Assignment step
+    for x in X:
+        clusters[argmin([((x - m)**2).sum() for m in M])].append(x)
+    # Update step
+    M = array([mean(c, 0) for c in clusters])
+    steps += 1
+
+print 'Completed in %d steps' % steps
+
+# Plot clusters
+figure(1)
+colors = [[1,0,0], [0,1,0], [0,0,1]]
+for i in range(k):
+    c = array(clusters[i])
+    scatter(c[:,0], c[:,1], c=colors[i])
+savefig('k-means.pdf')
+show()