Applying a function with arguments to pandas dataframe

Question

I have a percentile dataframe df as follows:

            col1   col2   col3      
GDStart                                                                   
2019-09-02  100    11     12  
2019-09-03   60    16     14  
2019-09-04   60    67     13

and an external function as:

import numpy as np
def Fn(x, a):
 return np.percentile(x, a)

I want to get apply my custom Fn to each row of df to get median for each row. I am expecting the following answer when using np.percentile(x, 50) i.e. median for each row:

         col1
GDStart                                                                   
2019-09-02   12
2019-09-03   16
2019-09-04   60

I am not sure how to apply the lambda or apply function here. My situation is tricky as I might need to find 30th percentile, or 50th percentile etc; a changes.

If you want `median` I suspect you meant to use `quantile` in `Fn` and not `percentile` — piRSquared, Jan 13 '20 at 17:22
Unless you're doing something else on top - will `df.median(axis=1)` not work here? — Jon Clements, Jan 13 '20 at 17:24
@JonClements I suspect they want to understand the mechanics of how to pass other arguments to the function being applied... which I'm looking for the dup now. — piRSquared, Jan 13 '20 at 17:25

score 2 · Accepted Answer · answered Jan 13 '20 at 17:20

2

Pass the keywords to `apply`

It will pass it along to the function

df.apply(Fn, a=.5, axis=1)

Pass the positional arguments

df.apply(Fn, args=(.5,), axis=1)

answered Jan 13 '20 at 17:20

piRSquared

285,575
57
475
624

Applying a function with arguments to pandas dataframe

1 Answers1

Pass the keywords to apply

Pass the positional arguments

Pass the keywords to `apply`