Pandas的Apply函数理解

 

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

import pandas as pd import datetime   #用来计算日期差的包   def dataInterval(data1,data2):     d1 = datetime.datetime.strptime(data1, '%Y-%m-%d')     d2 = datetime.datetime.strptime(data2, '%Y-%m-%d')     delta = d1 - d2     return delta.days   def getInterval(arrLike):  #用来计算日期间隔天数的调用的函数     PublishedTime = arrLike['PublishedTime']     ReceivedTime = arrLike['ReceivedTime'] #    print(PublishedTime.strip(),ReceivedTime.strip())     days = dataInterval(PublishedTime.strip(),ReceivedTime.strip())  #注意去掉两端空白     return days   if __name__ == '__main__':        fileName = "NS_new.xls";     df = pd.read_excel(fileName)     df['TimeInterval'] = df.apply(getInterval , axis = 1)

　　

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

import pandas as pd import datetime   #用来计算日期差的包   def dataInterval(data1,data2):     d1 = datetime.datetime.strptime(data1, '%Y-%m-%d')     d2 = datetime.datetime.strptime(data2, '%Y-%m-%d')     delta = d1 - d2     return delta.days   def getInterval_new(arrLike,before,after):  #用来计算日期间隔天数的调用的函数     before = arrLike[before]     after = arrLike[after] #    print(PublishedTime.strip(),ReceivedTime.strip())     days = dataInterval(after.strip(),before.strip())  #注意去掉两端空白     return days     if __name__ == '__main__':        fileName = "NS_new.xls";     df = pd.read_excel(fileName)     df['TimeInterval'] = df.apply(getInterval_new ,       axis = 1, args = ('ReceivedTime','PublishedTime'))    #调用方式一     #下面的调用方式等价于上面的调用方式     df['TimeInterval'] = df.apply(getInterval_new ,       axis = 1, **{'before':'ReceivedTime','after':'PublishedTime'})  #调用方式二     #下面的调用方式等价于上面的调用方式     df['TimeInterval'] = df.apply(getInterval_new ,       axis = 1, before='ReceivedTime',after='PublishedTime')  #调用方式三

　　

修改后的getInterval_new函数多了两个参数，这样我们在使用apply函数的时候要自己
传递参数，代码中显示的三种传递方式都行。

最后，本篇的全部代码在下面这个网页可以下载：

https://github.com/Dongzhixiao/Python_Exercise/tree/master/pandas_apply