Pandas merge rows

Question

I have a single dataframe like this:

df1 = pd.DataFrame([{'GameID': 15, 'Column1': 20,'Column2': 25, 'Column3': -15,'Column4': '','Column5': ''}, {'GameID': 15, 'Column1': '','Column2': '','Column3': '','Column4': 30,'Column5': 40}])

The output is like this:

   GameID Column1 Column2 Column3 Column4 Column5
0      15      20      25     -15
1      15                              30      40

What I'm tring to do here is to merge my row, meaning that the output will be like

   GameID Column1 Column2 Column3 Column4 Column5
0      15      20      25     -15      30      40

I've tried to groupby but the result isn't the one expected

df1 = pd.DataFrame(df1.groupby('GameID'))
    0                                                  1
0  15     GameID Column1 Column2 Column3 Column4 Colu...

Any suggestion on this topic will be much appreciated

Thanks

Geoffrey

Answer 1

This is another alternative without groupby() as greatly suggested in the comments, using np.nan and fillna() :

df1 = df1.replace({'':np.nan}).fillna(method='bfill').head(1)

Output:

GameID  Column1 Column2 Column3 Column4 Column5
0   15  20.0    25.0    -15.0   30.0    40.0