CODEWITHSUNDEEP
pythonpython-3.xpandasdataframe
mattylim
mattylim

Reputation: 13

How do I change the value in a dataframe's column based on another column?

I have a dataframe where I want to pass a function to a specific column:

enter image description here

Looking to change the 'Channel' column values based on certain requirements:

if 'Facebook' or 'FacebookPaid' or 'FacebookFree' in df['utm_Campaign']:
    df['Channel'] = 'FB'
elif 'Newsletter' in df['utm_Campaign']:
    df['Channel'] = 'Email'
else:
   if 'YoutubePaid' in df['utm_Campaign']:
                df['Channel'] = 'Youtube Direct'

Basically if there is a word 'Facebook' contained in another column (utm_Campaign) from the same row, the string in the 'Channel' column for the dataframe is set as 'FB'.

Not sure what to do. Do i map a function, or do a df.replace, or something else?

Upvotes: 1

Views: 105

Answers (3)

fabiob
fabiob

Reputation: 295

Just in case you wanted to use apply:

df['Channel'] = df.apply(lambda df: \
    'FB' if df['utm_Campaign'].str.contains('Facebook') else \
    ('Email' if df['utm_Campaign'].str.contains('Newsletter') else \
    ('Youtube Direct' if df['utm_Campaign'].str.contains('YoutubePaid'))))

Upvotes: 0

screenpaver
screenpaver

Reputation: 1130

You could also do something like this:

df['Channel'] = ['FB' if 'Face' in row else df.iloc[i].Channel for i,row in enumerate(df.utm_Campaign)]

Upvotes: 0

jezrael
jezrael

Reputation: 863741

You can create conditions by str.contains and then use numpy.select:

m1 = df['utm_Campaign'].str.contains('Facebook|FacebookPaid|FacebookFree')
m2 = df['utm_Campaign'].str.contains('Newsletter')
m3 = df['utm_Campaign'].str.contains('YoutubePaid')

df['Channel'] = np.select([m1,m2,m3], ['FB','Email','Youtube Direct'], default='no match')

Upvotes: 1

Related Questions