def hie_clustering(K, Idata):
agglomerative1 = AgglomerativeClustering(n_clusters=K,
linkage='ward',
affinity='euclidean')
agglomerative1.fit_predict(Idata)
hlabels = agglomerative1.labels_
#print "hlabels : ", hlabels
centers = cal_center(Idata, hlabels)
ss = silhouette_score(Idata, hlabels)
# di = calculate_dunn_index(Idata,label,n_cluster=i+1)
pbm = cal_pbm_index(K, Idata, centers, hlabels)
return centers, hlabels, ss, pbm
#
# Idata=np.array([[1,2,3],[3,2,1],[4,5,6],[6,5,4],[66,74,67],[74,67,66],[44,26,34],[967,589,122]])
K = 24
print "No of clusters : ", K
cluster = KMeans(n_clusters=K)
cluster.fit(Idata)
label = cluster.predict(Idata)
centers = cluster.cluster_centers_
#print "Kmeans center : ", centers
#print "cal centers : ", cal_center(Idata,label)
dn = calculate_dunn_index(Idata, label, n_cluster=K)
print "Kmeans Siloutte score : ", silhouette_score(Idata, label)
print "PBM index : ", cal_pbm_index(K, Idata, centers, label)
print "Dunn index : ", dn
print "-------------------------------------------------------------\n"
agglomerative1 = AgglomerativeClustering(n_clusters=K,
linkage='ward',
affinity='euclidean')
idx = agglomerative1.fit_predict(Idata)
hlabels = agglomerative1.labels_
centers = cal_center(Idata, hlabels)
dn = calculate_dunn_index(Idata, hlabels, n_cluster=K)
print "Agg ward Siloutte score : ", silhouette_score(Idata, hlabels)
print "PBM index : ", cal_pbm_index(K, Idata, centers, hlabels)
print "Dunn index : ", dn
print "---------------------------------------------------------\n"
# Give no. of clusters K = 2 # -------------------------------- cluster = KMeans(n_clusters=K) cluster.fit(Idata) label = cluster.predict(Idata) centers = cluster.cluster_centers_ dn = calculate_dunn_index(Idata, label, K) print "Kmeans Siloutte score : ", silhouette_score(Idata, label) print "Kmeans PBM index : ", cal_pbm_index(K, Idata, centers, label) print "Kmeans Dunn index : ", dn print "-------------------------------------------------------------\n" agglomerative1 = AgglomerativeClustering(n_clusters=K, linkage='ward', affinity='euclidean') idx = agglomerative1.fit_predict(Idata) hlabels = agglomerative1.labels_ centers = cal_center(Idata, hlabels) dn = calculate_dunn_index(Idata, hlabels, K) print "Hierarchical single Siloutte score : ", silhouette_score(Idata, hlabels) print "Hierarchical single PBM index : ", cal_pbm_index(K, Idata, centers, hlabels) print "Hierarchical single Dunn index : ", dn print "---------------------------------------------------------\n" agglomerative = AgglomerativeClustering(n_clusters=K, linkage='average', affinity='euclidean') agglomerative.fit_predict(Idata)
K.insert(i, n)
print "no. of cluster : ", n
cluster = KMeans(n_clusters=n)
cluster.fit(Idata)
label = cluster.predict(Idata)
centers = cluster.cluster_centers_
a = centers.tolist()
for j in range(len(a)):
for k in range(len(Idata[0])):
pop.append(a[j][k])
if not max_choromosome_length - len(pop) == 0:
extra_zero = max_choromosome_length - len(pop)
pop.extend(0 for x in range(extra_zero))
x.insert(i, pop)
ss = silhouette_score(Idata, label)
pbm = cal_pbm_index(n, Idata, centers, label)
sil_sco.insert(i, ss)
PBM.insert(i, pbm)
Initial_Label.insert(i, label.tolist())
Final_label = list(Initial_Label)
print "Chromosome clusers :", K
print "Chromosomes PBM Index :", PBM
print "Chromosome SI : ", sil_sco
print "Max PBM, Sil : ", max(PBM), ",", max(sil_sco)
print "clusters corresponding to Max PBM, Sil : ", K[PBM.index(
max(PBM))], ", ", K[sil_sco.index(max(sil_sco))]
# file2.write("\n Initial clusters : "+ str(K) + '\n' )
# file2.write("Chromosome initial PBM index: "+ str(PBM) + '\n' )
# file2.write("Chromosome initial Sil score : " +str(sil_sco)+ '\n')