pandas替换引用用户输入的值

使用numpy where 如下:

import numpy as np
import pandas as pd

# reproducible example
df = pd.DataFrame()
df['Area'] = ['Sales','Finance','Finance','Ops']
df['Name'] = ['Tom','Laura','An','Roger']
df['Category'] = [1,1,1,1]
d = {'finance':2 ,'sales':3 ,'ops':4}

selection = input("Please type in what areas you are searching for:")
# assumes the user is typing in multiple areas split by space
selection = selection.lower().split(" ")
# loop through each selection and change the category
for s in selection:
# takes care of searches that aren't in the dictionary
    if s in d.keys():
        df['Category'] = np.where(df['Area'].str.lower()==s,d[s],df['Category'])

例如,如果用户键入“ops Finance”,则输出为

    Area    Name    Category
0   Sales   Tom     1
1   Finance Laura   2
2   Finance An      2
3   Ops     Roger   4

您需要检查列表中的值是否是字典的一部分,如果是,您可以使用iloc选项查找该值并将其替换为新值。

以下是完整的代码:

c = ['Area','Name','Category']
d = [['Sales','Tom',1],
['Finance','Laura',1],
['Finance','An',1],
['Ops','Roger',1]]

dct = {'finance':'2' ,'sales':'3' ,'ops':'4'}

import pandas as pd
df = pd.DataFrame(d,columns=c)
print (df)

selection = ['Finance','Sales']

#the if statement will ensure selection = 'Finance' gets converted to a list
#that way the for loop is not impacted
if isinstance(selection,str): selection = [selection]

for cat in selection:
    if cat.lower() in dct:
        df.loc[df['Area'] == cat, 'Category'] = dct[cat.lower()]

print (df)

这些输出是:

Input : selection = 'Finance'

输出为:

      Area   Name Category
0    Sales    Tom        1
1  Finance  Laura        2
2  Finance     An        2
3      Ops  Roger        1

输入:选择=[“财务”,“销售”]

输出为:

      Area   Name Category
0    Sales    Tom        3
1  Finance  Laura        2
2  Finance     An        2
3      Ops  Roger        1

如果你决定通过输入语句从用户那里获取输入,那么你可以添加以下代码(并替换赋值):

替换此代码:

selection = ['Finance','Sales']
if isinstance(selection,str): selection = [selection]

使用此代码:

sel = input('What\'s the selection critiera :')
selection = [s.capitalize() for s in re.split(r' |,',sel) if s != '']