def runAff(fileName, title='AffinityPropagation'):
propertysMatrix, labelMatrix = loadData(fileName)
cluster = AffinityPropagation().fit(propertysMatrix)
myLabel = cluster.labels_
re = computeARI(myLabel, labelMatrix)
print re
drawLabel(propertysMatrix, myLabel, t=title)
def runDBSCAN(fileName,eps,min_samples,title='DBSCAN'):
propertysMatrix, labelMatrix = loadData(fileName)
cluster = DBSCAN(eps=eps, min_samples=min_samples,metric='euclidean').fit(propertysMatrix)
myLabel = cluster.labels_
re= computeARI(myLabel, labelMatrix),eps,min_samples
print re
# drawLabel(propertysMatrix, myLabel, t=title)
return re[0][-3]
data = []
for filename in fileList.keys():
X, labels_true = loadData(filename)
# drawLabel(X, LabelA=labels_true, )
# 原始聚类
# 其他算法最好的结果
k = len(set(labels_true))
ap = AffinityPropagation().fit(X)
kmeans = KMeans(n_clusters=k, random_state=0).fit(X)
hierarchical = AgglomerativeClustering(n_clusters=k).fit(X)
f = filename.split('/')[-1].split('.')[0]
dbscan = DBSCAN(eps=DBSCAN_para[f][0],
min_samples=DBSCAN_para[f][1],
metric='euclidean').fit(X)
re_dbscan = computeARI(dbscan.labels_, labels_true, X=X)
re_ap = computeARI(ap.labels_, labels_true, X=X)
re_kMeans = computeARI(kmeans.labels_, labels_true, X=X)
re_Hierarchical = computeARI(hierarchical.labels_, labels_true, X=X)
re_max = np.max(np.vstack(
(re_kMeans, re_Hierarchical, re_ap, re_dbscan)),
axis=0)
# ===============
noisePercent = computeNoise(SIA2WA(file2SIA(filename), percent=0.05))
if noisePercent < 0.05:
percent = 2 * noisePercent + 0.1
cblp = CBLP(model='even', percent=percent).fit(X)
re_cblp = computeARI(cblp.labels_, labels_true, X=X)
print('model:even,percent:', percent, re_cblp,
(re_max < computeARI(cblp.labels_, labels_true, X=X)) * 1)
def runSpectralClustering(fileName, k=2):
propertysMatrix, labelMatrix = loadData(fileName)
re = SpectralClustering(n_clusters=k).fit(propertysMatrix)
myLabel = re.labels_
print computeARI(myLabel, labelMatrix)
drawLabel(propertysMatrix, myLabel, t='SpectralClustering')
def runKMeans(fileName, k=2):
propertysMatrix, labelMatrix = loadData(fileName)
kmeans = KMeans(n_clusters=k, random_state=0).fit(propertysMatrix)
myLabel = kmeans.labels_
print computeARI(myLabel, labelMatrix)
for i in os.listdir(evenPath):
if i.split('.')[-1] == 'txt':
evenFileList.append(evenPath + i)
unevenPath = '/Users/liuqiang/Documents/标签传播过程论文/聚类数据集/uneven/'
unevenFileList = []
for i in os.listdir(unevenPath):
if i.split('.')[-1] == 'txt':
unevenFileList.append(unevenPath + i)
unevenFileList = ['/Users/liuqiang/Documents/标签传播过程论文/聚类数据集/uneven/cancer.txt',]
for filename in unevenFileList:
X, labels_true = loadData(filename)
k = len(set(labels_true))
kmeans = KMeans(n_clusters=k, random_state=0).fit(X)
hierarchical = AgglomerativeClustering(n_clusters=k).fit(X)
re_kMeans = computeARI(kmeans.labels_, labels_true, X=X)
re_Hierarchical = computeARI(hierarchical.labels_, labels_true, X=X)
re = np.vstack((re_kMeans, re_Hierarchical))
re_max = np.max(re, axis=0)
print re_max
for i in range(4, 100, 5):
cblp = CBLP(model='uneven', topN=i).fit(X)
print i,computeARI(cblp.labels_, labels_true, X=X)
print i, (re_max < computeARI(cblp.labels_, labels_true, X=X)) * 1
# for filename in evenFileList:
# X, labels_true = loadData(filename)
# k = len(set(labels_true))
# kmeans = KMeans(n_clusters=k, random_state=0).fit(X)
# hierarchical = AgglomerativeClustering(n_clusters=k).fit(X)
# re_kMeans = computeARI(kmeans.labels_, labels_true, X=X)
def runAgglomerative(fileName, k=2):
propertysMatrix, labelMatrix = loadData(fileName)
re = AgglomerativeClustering(n_clusters=k).fit(propertysMatrix)
myLabel = re.labels_
print computeARI(myLabel, labelMatrix)