How to assign non null values in one group to all rows in the group in pandas?

Question

I have a dataframe that looks like this.

+----+-------+
| ID | Value |
+----+-------+
|  1 | 23    |
|  1 | NA    |
|  1 | NA    |
|  1 | NA    |
|  2 | 24    |
|  2 | NA    |
|  2 | NA    |
+----+-------+

For each ID value in a group, I either have one value or NA. I wanted to apply this non NA value to the NA values in that specific group as below.

Desired Output :

+----+-------+
| ID | Value |
+----+-------+
|  1 | 23    |
|  1 | 23    |
|  1 | 23    |
|  1 | 23    |
|  2 | 24    |
|  2 | 24    |
|  2 | 24    |
+----+-------+

How can we achieve this in pandas?

This worked. Thanks. Just one change. df = df.sort_values(['ID', 'Value']).ffill() — Roshan Joe Vincent, Aug 23 '20 at 16:53

score 2 · Answer 1 · answered Aug 23 '20 at 16:35

2

Check with

df.Value.fillna(df.groupby('ID').Value.transform('first'), inplace=True)

answered Aug 23 '20 at 16:35

BENY

317,841
20
164
234

Heyy benyo, Good seeing you here. Thanks for your cumsum solution, it really did the trick. ~ And wishing you Happy Coding as well. – High-Octane Aug 23 '20 at 16:40
@High-Octane happy coding ~ – BENY Aug 23 '20 at 16:52

High-Octane · Answer 2 · 2020-08-23T16:40:34.133

1

say your dataframe variable is df

Then,

df.value.fillna(method='ffill')

Should do the trick.

Documentation for your reference :

Forward Fill Documentation Reference

edited Aug 23 '20 at 16:40

answered Aug 23 '20 at 16:35

High-Octane

1,104
5
19

thanks. RichieV's answer worked for me – Roshan Joe Vincent Aug 23 '20 at 16:46

How to assign non null values in one group to all rows in the group in pandas?

2 Answers2