准备了一个包含500多万条数据的txt文件,分别使用numpy的genfromtxt()方法和pandas的read_table()方法,对读取时间进行了计时,可以得出结论,在读取大量数据方面,panda比numpy要快一个数量级。
代码如下:
import numpy as np
import pandas as pd
import time
start_time = time.time()
data = np.genfromtxt('rating.txt',delimiter=',')
end_time = time.time()
print("Numpy reading time:"+str(round((end_time-start_time)*1000,2)))
start_time = time.time()
data = pd.read_table('rating.txt',delimiter=',')
end_time =time.time()
print("Pandas reading time:"+str(round((end_time-start_time)*1000,2)))
运行结果如下:
Numpy reading time:128914.37
Pandas reading time:8565.49
[Finished in 142.9s]