Pandas update multiple columns using apply function

Question

I am updating a data frame using apply of function.

But now I need to modify multiple columns using this function,

Here is my sample code:

def update_row(row):
    listy = [1,2,3]
    return listy

dp_data_df[['A', 'P','Y']] = dp_data_df.apply(update_row, axis=1)

It is throwing the following error:

ValueError: shape mismatch: value array of shape (10,) could not be broadcast to indexing result of shape (3,10)

Thanks in advance.

jezrael · Accepted Answer · 2018-08-06 08:41:40Z

21

You can return pd.Series:

dp_data_df = pd.DataFrame({'A':[3,5,6,6],
                           'B':[6,7,8,9],
                           'P':[5,6,7,0],
                           'Y':[1,2,3,4]})
print (dp_data_df)
   A  B  P  Y
0  3  6  5  1
1  5  7  6  2
2  6  8  7  3
3  6  9  0  4

def update_row(row):
    listy = [1,2,3]
    return pd.Series(listy)

dp_data_df[['A', 'P','Y']] = dp_data_df.apply(update_row, axis=1)
print (dp_data_df)
   A  B  P  Y
0  1  6  2  3
1  1  7  2  3
2  1  8  2  3
3  1  9  2  3

answered Aug 6, 2018 at 8:41

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

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

3 Comments

Ahsan Over a year ago

@jezrael is there a way that I could change column names for each record and still use apply?

jezrael Over a year ago

@Ahsan - Not sure if understand, how is possible change columns names?

Abu Shoeb Over a year ago

@jezrael how can I use a condition for the update? For example, I want to update the row if the col B = 8 or 9.

Itachi · Accepted Answer · 2019-08-20 09:00:33Z

3

You can use zip your output:

def update_row(row):
    listy = [1,2,3]
    return listy

dp_data_df['A'], dp_data_df['P'], dp_data_df['Y'] = zip(*dp_data_df.apply(update_row, axis=1))

answered Aug 20, 2019 at 9:00

Itachi

3,0351 gold badge32 silver badges36 bronze badges

2 Comments

gosuto Over a year ago

Very elegant solution! Do you mind explaining how it works?

dan Over a year ago

This solution avoids the numeric return and allows the return of the char function. It works fine and is what I was looking for, but I don't understand how it works.

Collectives™ on Stack Overflow

Pandas update multiple columns using apply function

2 Answers 2

3 Comments

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related