CODEWITHSUNDEEP
pythonpython-3.xpandasdata-science
Chris90
Chris90

Reputation: 1998

Creating new columns based on value of other column

In my df I have a column for each entity below (Grubhub, Toasttab, Tenk) and it says either yes or no in the value of that column for each row,

I have below code such as:

df['Grubhub'] = df[['On GrubHub or Seamless?']].apply(lambda x: any(x == 'Yes'), axis = 1)

df['ToastTab'] = df[['On ToastTab?']].apply(lambda x: any(x == 'Yes'), axis = 1)

df['Tenk'] = df[['On Tenk?']].apply(lambda x: any(x == 'Yes'), axis = 1)

df['Udemy'] = df[['On Udmey?']].apply(lambda x: any(x == 'Yes'), axis = 1)

df['Postmates'] = df[['On Postmates?']].apply(lambda x: any(x == 'Yes'), axis = 1)

df['Doordash'] = df[['On DoorDash?']].apply(lambda x: any(x == 'Yes'), axis = 1)

df['Google'] = df[['On Goole?']].apply(lambda x: any(x == 'Yes'), axis = 1)

This gives me a new column for each entity ( Grubhub, Toasttab, Tenk ) and that column gives a true of false value, is there a more efficient method where I can do all of these in one line of code or function? Thanks for help in advance

Upvotes: 1

Views: 97

Answers (1)

Danila Ganchar
Danila Ganchar

Reputation: 11242

You can create a column map and apply a function inside loop:

columns_map = (
    ('Grubhub', 'On GrubHub or Seamless?'),
    ('ToastTab', 'On ToastTab?'),
    ('Tenk', 'On Tenk?'),
    # etc ...
)

for new_col, alias in columns_map:
    df[new_col] = df[alias].apply(lambda x: x == 'Yes')
    # also you can easily remove aliases columns:
    # df = df.drop(columns=[alias])

Or you can set value into original column and rename if you need(without drop()):

for new_col, alias in columns_map:
    df[alias] = df[alias].apply(lambda x: x == 'Yes')

df.rename(
    columns={alias: new_col for new_col, alias in columns_map},
    inplace=True
)

Upvotes: 2

Related Questions