def kmeansTest(k=2, n=20, verbose=False):
random.seed(0)
xMean = 3
xSD = 1
yMean = 5
ySD = 1
d1Samples = util.genDistribution(xMean, xSD, yMean, ySD, n, '1.')
d2Samples = util.genDistribution(xMean+3, xSD, yMean+1, ySD, n, '2.')
allSamples = d1Samples + d2Samples
print("before clustering")
util.plot_cluster([cluster.Cluster(allSamples)])
print("after clustering")
clusters = kmeans(allSamples, k, verbose)
util.plot_cluster(clusters, verbose)
print('Final result')
for c in clusters:
print('', c)
def make_data(n, scale=1):
""" A simple y = x curve, with noisy displacement on both
both x and y axis; change scale to change the range
"""
linear_data = [
sample.Sample('',
[float(x) / scale, float(x) / scale], '')
for x in range(n)
]
noise = util.genDistribution(xSD=0.3, ySD=0.3, n=n)
data = [linear_data[i] + noise[i] for i in range(n)]
return data
def make_data(n):
C = [random.choice(LABELS) for x in range(n)]
linear_data = [
sample.Sample(C[x], [x / (float(SCALE)), x / (float(SCALE))], C[x])
for x in range(n)
]
mean = 0
std = DEV * SCALE
noise = util.genDistribution(mean, std, mean, std, n, '')
data = [linear_data[i] + noise[i] for i in range(n)]
return data
#### Implement the centroid updating function here!
def update(self, samples):
"""Replace the samples in the cluster by new samples
Return: how much the centroid has changed"""
return helper.update(self, samples)
def __str__(self):
names = []
for e in self.samples:
names.append(e.getName())
names.sort()
result = 'Cluster with centroid '\
+ str(self.centroid.getFeatures()) + ' contains:\n '
for e in names:
result = result + e + ', '
return result[:-2]
if __name__ == "__main__":
test_samples = util.genDistribution()
c = Cluster(test_samples)
print(c.centroid)
print("cluster center: ", c.centroid.features)
util.plot_cluster([c])
# now assign the cluster new samples, and move it
test_samples2 = util.genDistribution(1, 1, 1, 1, 30)
diff = c.update(test_samples2)
print("center moved: ", diff)
# plot_cluster expects an array of cluster...
util.plot_cluster([c])
print(p)
if __name__ == "__main__":
# make data
random.seed(0)
n = 100
K = 3
LABELS = ('a', 'b', 'c')
all_cluster = []
data = []
for i in range(K):
tmp_data = util.genDistribution(i * 2 + 1,
1,
i * 2 + 1,
1,
n=20,
label=LABELS[i])
all_cluster.append(cl.Cluster(tmp_data))
data += tmp_data
def onclick(event):
# Creating a new point and finding the k nearest neighbours
new = sample.Sample('', [event.xdata, event.ydata], '')
knn(new, data, K)
# draw the new point
data.append(new)
pylab.scatter([new.getFeatures()[0]], \
[new.getFeatures()[1]], \
label = new.getLabel(), \
"""
max_label = util.LABELS[0]
p.setLabel(max_label)
# above forces a fixed label: remove them
# replace knn_helper.knn(p, data, k) with your own logic
print(p)
knn_helper.knn(p, data, k)
print(p)
if __name__ == "__main__":
random.seed(0)
n = 100
K = 3
data = util.genDistribution(n=10)
for d in data:
d.setLabel(random.choice(util.LABELS))
print("before....")
util.plot_data(data)
new_pt = sample.Sample('', [0.2, 0.3], '')
knn(new_pt, data, K)
data.append(new_pt)
print("\nafter....")
util.plot_data(data)