Pandas Dataframe Series : check if specific value exists

Question

I need to iterate over a list and perform a specific operation if the value from the list exists in one of the pandas dataframe column. I tried to do as below, but getting below error

'Error: #The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().'

import pandas as pd

people = {
    'fname':['Alex','Jane','John'],
    'age':[20,15,25],
    'sal':[100,200,300]
}

df=pd.DataFrame(people)

check_list=['Alex','John']

for column in check_list:
    if (column == df['fname']):
        df['new_column']=df['sal']/df['age']
    else:
        df['new_column']=df['sal']

df

Required output:

fname   age sal new_column
Alex    20  100  5      <<-- sal/age
Jane    15  200  200    <<-- sal as it is
John    25  300  12     <<-- sal/age

Umar.H · Accepted Answer · 2020-05-20T15:28:48.447

6

use np.where with .isin to check if a column contains particular values.

df['new_column'] = np.where(
        df['fname'].isin(['Alex','John']),
        df['sal']/df['age'],
        df['sal']
)

print(df)

  fname  age  sal  new_column
0  Alex   20  100         5.0
1  Jane   15  200       200.0
2  John   25  300        12.0

pure pandas version.

df['new_column'] = (df['sal']/df['age']).where(
                            df['fname'].isin(['Alex','John']),other=df['sal'])

print(df)
 fname  age  sal  new_col
0  Alex   20  100      5.0
1  Jane   15  200    200.0
2  John   25  300     12.0

edited May 20 '20 at 15:28

answered May 20 '20 at 15:18

Umar.H

22,559
7
39
74

Thanks Datanovice. Is there any other way to do this without using np.where? – steve May 20 '20 at 15:24
@steve see edit. – Umar.H May 20 '20 at 15:28

score 2 · Answer 2 · answered May 20 '20 at 15:22

Try using df.apply

import pandas as pd

people = {
    'fname':['Alex','Jane','John'],
    'age':[20,15,25],
    'sal':[100,200,300]
}

df=pd.DataFrame(people)

def checker(item):
    check_list=['Alex','John']
    if item["fname"] in check_list:
        return item['sal']/item['age']
    else:
        return item['sal']

df["Exists"] = df.apply(checker, axis=1)

df

score 1 · Answer 3 · answered May 20 '20 at 15:26

for index,row in df.iterrows():
    if row['fname'] in check_list:
           df.at[index,'new_column']=row['sal']/row['age']
    else:
           df.at[index,'new_column']=row['sal']

explanation: To loop over dataframe, use iterrows(), row variable will have values for all columns, index is the index of the row.

Pandas Dataframe Series : check if specific value exists

3 Answers3