Created
June 14, 2018 00:08
-
-
Save nekketsuuu/d2d4e565018cf44b3898c4d6208eece8 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import numpy as np | |
import matplotlib.pyplot as plt | |
from sklearn.cluster import KMeans | |
class GlobalKmeans(object): | |
def __init__(self, X, cluster): | |
self.X = X | |
self.num = X.shape[0] | |
self.label = np.zeros((self.num, 1)) | |
self.cluster = cluster | |
self.k = 0 | |
self.centroid = np.zeros((self.cluster, self.X.shape[1])) | |
def distance(self, x, y): | |
return np.linalg.norm(x-y) | |
def assign(self): | |
min_inertia = 0.0 | |
new_centroids = np.zeros((self.k, self.X.shape[1])) | |
for i in xrange(self.num): | |
cs = np.concatenate((self.centroid, self.X[i].reshape((1, 2))), axis=0) | |
kmeans = KMeans(n_clusters=self.k, init=cs, n_init=1).fit(self.X) | |
if i == 0 or min_inertia > kmeans.inertia_: | |
min_inertia = kmeans.inertia_ | |
new_centroids = kmeans.cluster_centers_ | |
self.centroid = new_centroids | |
def give_label(self): | |
for i in xrange(self.num): | |
temp = self.distance(self.X[i], self.centroid[0]) | |
labe = 0 | |
for k in xrange(self.k): | |
if k == 0: | |
continue | |
d = self.distance(self.X[i], self.centroid[k]) | |
if temp > d: | |
labe = k | |
temp = d | |
self.label[i] = labe | |
def fit(self): | |
self.k = 1 | |
self.centroid = np.mean(self.X, axis = 0).reshape((1,2)) | |
while(self.k < self.cluster): | |
self.k = self.k + 1 | |
self.assign() | |
self.give_label() | |
def plot(self): | |
plt.figure(figsize=(10,10)) | |
for j in xrange(self.cluster): | |
points = X[np.where(self.label == j)[0]] | |
plt.scatter(points[:,0], points[:,1]) | |
plt.scatter(self.centroid[j,0], self.centroid[j,1], s = 300, marker = "x") | |
if __name__ == '__main__': | |
N = 30 | |
mu1 = [-10, -15] | |
sigma1 = [[1, 0], [0, 1]] | |
mu2 = [0, 1] | |
sigma2 = [[1, 0], [0, 1]] | |
mu3 = [3, 5] | |
sigma3 = [[1, 0], [0, 1]] | |
x1 = 0.2 * np.random.multivariate_normal(mu1, sigma1, N) | |
x2 = 0.2 * np.random.multivariate_normal(mu2, sigma2, N) | |
x3 = 0.2 * np.random.multivariate_normal(mu3, sigma3, N) | |
X = np.concatenate([x1, x2, x3], axis=0) | |
g = GlobalKmeans(X, 3) | |
g.fit() | |
g.plot() | |
plt.show() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Thank you very much!!