python kmeans

sklearn官网documentation

http://scikit-learn.org/dev/modules/generated/sklearn.cluster.KMeans.html

# K-Means Algorithm
import numpy as np
from sklearn.cluster import KMeans

def loadData(file):
    tmp = np.loadtxt(file, dtype=np.str, delimiter=",")
    data = tmp[0:1514, 0:5].astype(np.float)  # load data
    return data

if __name__ == '__main__':
    handledData = loadData("handled300001.CSV")
    estimator = KMeans(n_clusters=2)
    estimator.fit_predict(handledData)# compute k-means clustering
    prelabel = estimator.labels_# labels
    centroids = estimator.cluster_centers_# centroids
    # inertia = estimator.inertia_# centroid sum

    res1 = []

    res2 = []# save the two component clusterings

    for items in range(len(handledData)):
        if (int(prelabel[items]) == 0):
            res1.append(handledData[items][0], handledData[items][1], handledData[items][2],\
                        handledData[items][3], handledData[items][4])
        elif (int(prelabel[items]) == 1):
            res2.append(handledData[items][0], handledData[items][1], handledData[items][2],\
                        handledData[items][3], handledData[items][4])

    print(prelabel, '\n', centroids)

我看官网的document，貌似sklearn的kmeans不能输出不能直接输出各个cluster 就补充了后面的res1和res2来分别存储两个cluster