CODEWITHSUNDEEP
pythonpandasdataframedictionaryconditional-statements
s_khan92
s_khan92

Reputation: 979

How to use multiple conditions based on 2 columns to create the new column in pandas?

I have 2 dictionaries and i need to create the column based on several condition.

dict1 = {'100': BMW, '200': Audi, '300': 'VW'}   
dict2 = {'100': Mercedes, '200': Nissan, '300': 'Renault'}

df:

class          Code
1              200
1              300
2              300
1              100
2              100

I actually want to use dict1 when class is 1 and dict2 when class is 2

desired output would be like this:

  class            Code         Car
    1              200          Audi
    1              300          VW
    2              300          Renault
    1              100          BMW
    2              100          Mercedes

I could use .map if i had no condition but i am not sure what to use now:

df['Car'] = df['Code'].map(dict1)

Upvotes: 0

Views: 556

Answers (2)

safiqul islam
safiqul islam

Reputation: 650

i tested with the follwing code

import pandas as pd
dict1 = {'100': 'BMW', '200': 'Audi', '300': 'VW'}
dict2 = {'100': 'Mercedes', '200': 'Nissan', '300': 'Renault'}
df = pd.DataFrame({'Class':[1,1,2,1],'Code':['200','300','300','100']})
def f(row):
    if row['Class'] == 1:
        val = dict1[row['Code']]
    elif row['Class'] ==2:
        val = dict2[row['Code']]
    else:
        val = dict2[row['Code']]
    return val

df['Car']= df.apply(f,axis=1)

print(df)

it prints

    Class Code     Car
0      1  200     Audi
1      1  300       VW
2      2  300  Renault
3      1  100      BMW

Upvotes: 1

Quang Hoang
Quang Hoang

Reputation: 150735

If it's just two dictionaries/classes:

# note that your dictionary has string key
df['Code'] = df.Code.astype(str)

df['car'] = np.where(df['class']==1,
                     df['Code'].map(dict1),
                     df['Code'].map(dict2) )

Output:

   class Code       Car
0      1  200      Audi
1      1  300        VW
2      2  300   Renault
3      1  100       BMW
4      2  100  Mercedes

Upvotes: 0

Related Questions