Scikit-Learn 演算法 分群

woff

程式碼

1. import numpy as np
   
2. import matplotlib.pyplot as plt
   
3. import pandas as pd
   
4. from sklearn import datasets
   
5. 
   
6. iris=datasets.load_iris() #下載資料
   
7. X=iris.data
   
8. X=X[:,2:4]
   
9. 
   
10. from sklearn.cluster import AgglomerativeClustering
    
11. #Sklearn
    
12. # n_clusters: 要分成幾群 ，給定數值
    
13. # affinity: 距離的計算方式，"euclidean","l1","l2","manhattan","cosine"...¶
    
14. # linkage: 群與群之間的距離，"ward","complete","average","single"
    
15. ml=AgglomerativeClustering(n_clusters=3,affinity='euclidean',linkage='ward')
    
16. ml.fit_predict(X)
    
17. plt.scatter(X[:,0],X[:,1],c=ml.fit_predict(X))
    
18. plt.show()

複製代碼

輸出 array
array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,       1, 1, 1, 1, 1, 1, 2, 2, 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,       2, 2, 2, 2, 0, 2, 0, 2, 2, 2, 2, 0, 2, 2, 2, 2, 2, 0, 2, 2, 2, 2,       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0,       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], dtype=int64)



程式碼

1. # Scipy
   
2. # metric: 距離的計算方式
   
3. # method: 群與群之間的計算方式，"single", "complete", "average", "weighted", "centroid", "median", "ward"
   
4. import scipy.cluster.hierarchy as sch
   
5. dis=sch.linkage(X,metric='euclidean',method='ward')
   
6. sch.dendrogram(dis)
   
7. plt.title('Hierarchical Clustering')
   
8. plt.show()

複製代碼

程式碼

1. max_dis=5
   
2. clusters=sch.fcluster(dis,max_dis,criterion='distance')
   
3. clusters
   
4. print('clusters',clusters)
   
5. plt.scatter(X[:,0],X[:,1],c=clusters)
   
6. plt.show()

複製代碼

程式碼

1. k=5
   
2. clusters=sch.fcluster(dis,k,criterion='maxclust')
   
3. plt.scatter(X[:,0],X[:,1],c=clusters)
   
4. plt.show()

複製代碼

來源