Python - Drop duplicate based on max value of a column
You need DataFrameGroupBy.idxmax for indexes of max value of value3 and thes select DataFrame by loc:
print (df.groupby(['id1','id2','value1']).value3.idxmax())id1 id2 value11 2 30 13 5 12 424 12 1 6Name: value3, dtype: int64df = df.loc[df.groupby(['id1','id2','value1']).value3.idxmax()]print (df) id1 id2 value1 value2 value3 a1 1 2 30 42 26.2 NaN4 3 5 12 33 11.2 NaN6 24 12 1 23 1.9 NaNAnother possible solution is sort_values by column value3 and then groupby with GroupBy.first:
df = df.sort_values('value3', ascending=False) .groupby(['id1','id2','value1'], sort=False) .first() .reset_index()print (df) id1 id2 value1 value2 value3 a0 1 2 30 42 26.2 NaN1 3 5 12 33 11.2 NaN2 24 12 1 23 1.9 NaN