-
Notifications
You must be signed in to change notification settings - Fork 0
/
k-means.py
39 lines (34 loc) · 1.14 KB
/
k-means.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
from sklearn.cluster import KMeans
import numpy as np
import matplotlib.pyplot as plt
import pickle
if __name__ == "__main__":
loaded = np.load('blue_points_indicies.npz')
x = loaded["x"]
y = loaded["y"]
X = np.stack((x, y), axis=1)
'''
kmeans_models = []
scores = []
num_clusters = [5,10,20,50,100,150,200]
for i in range(len(num_clusters)):
print(i)
kmeans_models.append(KMeans(n_clusters=num_clusters[i], random_state=0).fit(X))
scores.append(kmeans_models[i].score(X))
# plt.figure(1)
plt.plot(num_clusters, scores)
plt.show() # around 20 is the best number of clusters '''
'''
for i in range(20,26):
kmeans = KMeans(n_clusters=i, random_state=0).fit(X)
plt.figure(i)
plt.title("K-Means Clusters: " + str(i))
plt.plot(x,y, "o", markersize=0.5)
plt.plot(kmeans.cluster_centers_[:,0], kmeans.cluster_centers_[:,1], "o", markersize=5)
plt.show()
kmeans_models.append(kmeans) '''
kmeans = KMeans(n_clusters=22, random_state=0).fit(X)
plt.plot(x,y, "o", markersize=0.5)
plt.plot(kmeans.cluster_centers_[:,0], kmeans.cluster_centers_[:,1], "o", markersize = 5)
plt.show()
pickle.dump(kmeans, open("kmeans22.sav", 'wb'))