Applying an if else function to two string columns in python / pandas

Question

In my dataframe I have:

Name    Sex    Height
Jackie   F       Small
John     M       Tall

I have made the following function to apply to create a new column based off combinations:

def genderfunc(x,y):
    if x =='Tall' & y=='M':
        return 'T Male'
    elif x =='Medium' & y=='M':
        return 'Male'
    elif x =='Small' & y=='M':
        return 'Male'
    elif x =='Tall' & y=='F':
        return 'T Female'
    elif x =='Medium' & y=='F':
        return 'Female'
    elif x =='Small' & y=='F':
        return 'Female'
    else:
        return y

My line of code to apply this function:

df['GenderDetails'] = df.apply(genderfunc(df['Height'],df['Sex']))

and i get the following:

TypeError: Cannot perform 'rand_' with a dtyped [object] array and scalar of type [bool]

Any ideas on what im doing wrong here? this is my first go at using a function.

Thanks!

sushanth · Accepted Answer · 2021-04-27 12:50:09Z

6

Here is another approach, using map.

map_ = {"TallM": "T Male", "SmallF": "Female"}

df['GenderDetails'] = (df['Height'] + df['Sex']).str.strip().map(map_)

     Name Sex Height GenderDetails
0  Jackie   F  Small        Female
1    John   M   Tall        T Male

answered Apr 27, 2021 at 12:50

sushanth

8,2923 gold badges20 silver badges31 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Nk03 · Accepted Answer · 2021-04-27 13:08:00Z

5

or you can use np.select, if performance is a concern-

condlist = [(df['Height'] == 'Tall') & (df['Sex'] == 'M'),
            (df['Height'] == 'Medium') & (df['Sex'] == 'M'),
            (df['Height'] == 'Small') & (df['Sex'] == 'M'),
            (df['Height'] == 'Tall') & (df['Sex'] == 'F'),
            (df['Height'] == 'Medium') & (df['Sex'] == 'F'),
            (df['Height'] == 'Small') & (df['Sex'] == 'F')]
choiselist = [
    'T Male',
    'Male',
    'Male',
    'T Female',
    'Female',
    'Female'
]

df['GenderDetails'] = np.select(condlist, choiselist, df['Sex'])

edited Apr 27, 2021 at 13:08

answered Apr 27, 2021 at 12:55

Nk03

15k2 gold badges11 silver badges24 bronze badges

Comments

Jacob Lee · Accepted Answer · 2021-04-28 02:09:36Z

2

You need to replace & with and.

edited Apr 28, 2021 at 2:09

Jacob Lee

4,7302 gold badges20 silver badges40 bronze badges

answered Apr 27, 2021 at 12:47

Shahriyar Mustafa

788 bronze badges

1 Comment

hfontanez Over a year ago

@pythonic833 I completely disagree with you. The EXACT recommendation was followed by the selected answer. The only difference is that the selected answer had the corrected code. Nevertheless, the answer here is correct.

jezrael · Accepted Answer · 2021-04-27 12:51:31Z

You are close, need lambda function with axis=1 and because scalar processing use and:

def genderfunc(x,y):
    if x =='Tall' and y=='M':
        return 'T Male'
    elif x =='Medium' and y=='M':
        return 'Male'
    elif x =='Small' and y=='M':
        return 'Male'
    elif x =='Tall' and y=='F':
        return 'T Female'
    elif x =='Medium' and y=='F':
        return 'Female'
    elif x =='Small' and y=='F':
        return 'Female'
    else:
        return y

df['GenderDetails'] = df.apply(lambda x: genderfunc(x['Height'],x['Sex']), axis=1)
print (df)
     Name Sex Height GenderDetails
0  Jackie   F  Small        Female
1    John   M   Tall        T Male

Non loop solution is possible with helper DataFrame and left join:

#set values like need
L = [('Tall','M','T Male'), ('Small','F','Female')]
df1 = pd.DataFrame(L, columns=['Height','Sex','GenderDetails'])


df = df.merge(df1, how='left')
print (df)
     Name Sex Height GenderDetails
0  Jackie   F  Small        Female
1    John   M   Tall        T Male

Collectives™ on Stack Overflow

Applying an if else function to two string columns in python / pandas

4 Answers 4

Comments

Comments

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

Comments

Comments

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related