Custom methods for a Pandas DataFrame

Question

I'd like to write custom functions that I can call on my pd.DataFrame df using the df.method() notation.

For instance,

def my_pd_method(df: pd.DataFrame, col: str)->pd.DataFrame:

    '''apply my_function to df[col] of df'''

    df_copy = df.copy(deep = True)
    df_copy[col] = df_copy[col].apply(lambda x: my_function(x), axis = 1)
    
   return df_copy

After this, I can run the command

PandasObject.my_pd_method = my_pd_method

to define the my_pd_method as a pd method.

After this, df.my_pd_method(col) will run as expected.

Is there some way to do this in a single function that I can put it in a library, import it and start using it, without having to run PandasObject.my_pd_method = my_pd_method?

vianmixt · Accepted Answer · 2022-11-21 14:32:13Z

2

Your best shot is to use inheritance i.e to create your own custom class that inherits from pandas DataFrame class.

Example:

class CustomDataFrame(pd.DataFrame):
    def my_method(self, col):
        df_copy = self.copy(deep = True)
        df_copy[col] = df_copy[col].apply(lambda x: my_function(x), axis = 1)

    return df_copy

Then you will be able to call your method like you wanted:

df.my_method(col)

edited Nov 21, 2022 at 14:32

answered Nov 18, 2022 at 8:08

vianmixt

9232 gold badges8 silver badges19 bronze badges

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

1 Comment

zab Over a year ago

Very nice. The last line of the function return df_copy fell out of the code snippet.

Collectives™ on Stack Overflow

Custom methods for a Pandas DataFrame

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related