Ignore minimum values from distincts rows in pandas

Question

I've below pandas dataframe

Name    |  Value1 |  Value2
User 1  |  1      |  15
User 1  |  2      |  35
User 1  |  3      |  35
User 1  |  3      |  35
User 2  |  4      |  23
User 2  |  5      |  25
User 3  |  6      |  45
User 3  |  7      |  50

I want to select only those rows which have largest value in Value2 column using pandas dataframe

Like this:

Name    |  Value1 |  Value2
User 1  |  2      |  35
User 1  |  3      |  35
User 1  |  3      |  35
User 2  |  5      |  25
User 3  |  7      |  50

Code with description will help me more as pandas are new for me.

Thanks in Advance :)

Answer 1

Use GroupBy.transform with max for Series filled by max values per groups, compare by column Value2 with Series.eq and filter by boolean indexing:

df1 = df[df.groupby('Name')['Value2'].transform('max').eq(df['Value2'])]
print (df1)

     Name  Value1  Value2
1  User 1       2      35
2  User 1       3      35
3  User 1       3      35
5  User 2       5      25
7  User 3       7      50

Details:

print (df.groupby('Name')['Value2'].transform('max'))
0    35
1    35
2    35
3    35
4    25
5    25
6    50
7    50
Name: Value2, dtype: int64

print (df.groupby('Name')['Value2'].transform('max').eq(df['Value2']))
0    False
1     True
2     True
3     True
4    False
5     True
6    False
7     True
Name: Value2, dtype: bool