[FIXED] How to conditionally rearrange row order

Issue

I have a dataset that has 'unknown' in each group. I am thinking to put 'unknown' at the end of each group. For example: I have a code like this:

import pandas as pd
clients1 = {'Name': ['A','B','Unknown','Y','Z','A','B','Unknown','Y','Z'],
        'group': ['1','1','1','1','1','2','2','2','2','2']
       }
df = pd.DataFrame(clients1, columns= ['Name', 'group'])
df

the output:

   Name group
    A   1
    B   1
    Unknown 1
    Y   1
    Z   1
    A   2
    B   2
    Unknown 2
    Y   2
    Z   2

the desired output should be:

   Name group
    A   1
    B   1
    Y   1
    Z   1
    Unknown 1
    A   2
    B   2
    Y   2
    Z   2
    Unknown 2

Does anyone know how to do that?

Solution

You can sort the data by group and name==Unknown, so the Unknown's are marked with True and put at the bottom:

df = (df.assign(unknown_name=df['Name'].eq('Unknown'))
        .sort_values(['group','unknown_name'])
     )

Output:

      Name group  unknown_name
0        A     1         False
1        B     1         False
3        Y     1         False
4        Z     1         False
2  Unknown     1          True
5        A     2         False
6        B     2         False
8        Y     2         False
9        Z     2         False
7  Unknown     2          True

Answered By - Quang Hoang