def split_data_plot(filename):
plot_data = []
gps_data,timestamp = readdata.get_network_data(filename)
for i in range(0,len(gps_data) - len(gps_data) % 30,30):
temp = numpy.array(gps_data[i:i + 30])
temp = temp.flatten()
temp = list(temp)
plot_data.append(temp)
return plot_data
filenames = os.listdir(path)
for name in filenames:
current = split_data_plot(path + '\\' + name)
plot = plot + current
gps_num.append(len(current))
return plot,gps_num
def calculate_pca(data):
p = sklearn.decomposition.pca.PCA(n_components = 5)
p.fit(data)
feature = p.transform(data)
return feature
if __name__ == '__main__':
# '''
gps_data,timestamp = readdata.get_network_data('.\\ubiqlog\\log_5-21-2014.txt')
# gps_data,timestamp = readdata.get_network_data('.\\log_10-31-2014.txt')
print len(gps_data)
# find_N_num(gps_data)
# cluster_centers,labels = k_means_cluster(gps_data,n_clusters = 2)
# labels = DBSCANJoint.dbscan_joint(gps_data,0.000005,30)
cluster_centers,labels = science_cluster(gps_data,30,show = False)
# draw_result(gps_data,labels)
print cluster_centers
'''
f = open('labels.txt','w')
for i in range(len(labels)):
f.write(str(i + 1))
f.write('\t')
f.write(str(labels[i]))
f.write('\n')
