Fill missing value

Question

I have a data sets which has column with missing value 2439. But the missing value is such that for specific index has some missing value and some fill value as shown below (Compare column 'Item_Identifier' and 'Item_Weight')

If carefully seen for specific item_identifier, there missing value in item_weight. Like this there many more Item_Identifier which missing value. Is there any way using python we fill missing value for only item_weight as same.

Possible duplicate of [How can I check for NaN in Python?](https://stackoverflow.com/questions/944700/how-can-i-check-for-nan-in-python) — pavi2410, Aug 04 '18 at 14:28

lintang · Accepted Answer · 2018-08-04T14:24:55.563

0

You can make the table into a pandas DataFrame then df['item_weight'].fillna(15.5, inplace=True)

edited Aug 04 '18 at 14:24

answered Aug 04 '18 at 14:22

lintang

100
5

1

He doesn't want it to be 0 – rafaelc Aug 04 '18 at 14:23
Sorry, he can fill it with any value. – lintang Aug 04 '18 at 14:24
1

What if OP has 100 columns? He will look for each value manually and use this? – rafaelc Aug 04 '18 at 14:26
what should be solution when there are more then one type of 'item_weight' and each have some missing value? – Kedar17 Aug 05 '18 at 06:58

score 0 · Answer 2 · answered Aug 04 '18 at 14:28

Reproducible example:

df = pd.DataFrame({'col1': ['a', 'a', 'b','b', 'b', 'c'], 
                   'col2': [10, np.nan, np.nan, np.nan, 20, 30]})

    col1    col2
0   a       10.0
1   a       NaN
2   b       NaN
3   b       NaN
4   b       20.0
5   c       30.0

You can groupby your col1 and agg using first

vals = df.groupby('col1').agg('first')

    col2
col1    
a   10.0
b   20.0
c   30.0

Then just use same indexing and fillna() to match and fill the values

df = df.set_index('col1').fillna(vals).reset_index()

    col1    col2
0   a       10.0
1   a       10.0
2   b       20.0
3   b       20.0
4   b       20.0
5   c       30.0

Fill missing value

2 Answers2