Question

我正在尝试创建一个函数，该函数将根据其他列的条件创建新列。当我只传递一个变量时，该函数工作正常，但当需要两个变量时，它不起作用。我想要做的一个例子是：

### create a function called name

def name(ID, NAME):
    if (ID == 1 ):
        return "First"
   elif (ID == 2):
        return "Second"
   elif (ID == 3):
       return "Third"
   elif (ID == 4 and NAME = “Four” ):
        return "Fourth"

### apply function to dataset and view results

dataset["NAME"].apply(name).head(100)

我的大多数新列值只能通过查看一个变量来获得，但少数需要两个。任何人都可以提供如何在python中实现这一目标的方向吗？在R中我使用了dplyr中的case_when函数，但我似乎没有发现python支持case语句

Answer 1

您可以将数据帧的整行传递给apply中带有axis = 1参数的函数，然后您可以访问函数中的部分行，如下所示：

import pandas as pd
import numpy as np

def nameme(row):
    if (row.ID == 1 ):
        return "First"
    elif (row.ID == 2):
        return "Second"
    elif (row.ID == 3):
       return "Third"
    elif (row.ID == 4 and row.Name == 'Four' ):
        return "Fourth"


dataset = pd.DataFrame({'ID':[0,1,2,3,4,5],'Name':['Four']*6})

dataset.apply(nameme, axis=1)

输出：

0      None
1     First
2    Second
3     Third
4    Fourth
5      None
dtype: object

Answer 2

我修改你的功能并创建一个玩具数据

def name(ID, NAME):
    if ID == 1 :
        return "First"
    elif ID == 2:
        return "Second"
    elif ID == 3:
        return "Third"
    elif ID == 4 and NAME == "Four" :
        return "Fourth"

dataset=pd.DataFrame({'ID':[1,2,3,4,4],'NAME':[1,2,3,4,'Four']})

dataset.apply(lambda x: name(x['ID'], x['NAME']), axis=1)

Out[741]: 
0     First
1    Second
2     Third
3      None# return None cause , it did not match all the condition 
4    Fourth
dtype: object

Answer 3

温家宝和斯科特波士顿有两个很好的答案。以下是我处理您DataFrame未必拥有您正在寻找的专栏的情况。因此，不要向您抛出错误，而是返回None：

def name(df):
    ID = df.get('ID')  # returns None if your DataFrame doesn't contain an 'ID' column
    NAME = df.get('NAME')  # returns None if your DataFrame doesn't contain a 'NAME' column
    if (ID == 1 ):
        return "First"
    elif (ID == 2):
        return "Second"
    elif (ID == 3):
        return "Third"
    elif (ID == 4 and NAME == "Four" ):
        return "Fourth"

data = pd.DataFrame({'ID':[1, 2, 3, 4, 4, 5], 'NAME':[1, 2, 3, 4, 'Four', 'Four']})
data['RESULT'] = data.apply(name, axis=1)

#    ID  NAME  RESULT
# 0   1     1   First
# 1   2     2  Second
# 2   3     3   Third
# 3   4     4    None
# 4   4  Four  Fourth
# 5   5  Four    None

当只有一些变量需要多个变量时，如何将多个值传递给函数-python 3

3 个答案: