如何根据出现排列 Dataframe 值

Question

Suppose I have a Dataframe having the following column values:

Name    |   Subject     |   Mark
------------------------------------
Daniel  |   Maths       |   95
Sam     |   Science     |   98
Nathan  |   English     |   90
Hobbs   |   Social      |   85
Shaw    |   Language    |   90
Daniel  |   Social      |   99
Shaw    |   Science     |   75
Nathan  |   Maths       |   99
Sam     |   Language    |   70
Hobbs   |   Language    |   90
Shaw    |   Social      |   90
Nathan  |   Social      |   85
Daniel  |   English     |   90
Nathan  |   Science     |   85
Hobbs   |   English     |   85
Nathan  |   Language    |   90
Daniel  |   Science     |   98
Sam     |   Social      |   85
Shaw    |   Maths       |   95
Daniel  |   Language    |   95
Sam     |   Maths       |   99
Hobbs   |   Science     |   99
Sam     |   English     |   75
Shaw    |   English     |   90
Hobbs   |   Maths       |   85

I need to create a code which would transform the dataframe into the following:

Name    |   Subject 1   |   Mark 1  |   Subject 2   |   Mark 2  |   Subject 3   |   Mark 3  |   Subject 4   |   Mark 4  |   Subject 5   |   Mark 5
----------------------------------------------------------------------------------------------------------------------------------------------------
Daniel  |   Maths       |   95      |   Social      |   99      |   English     |   90      |   Science     |   98      |   Language    |   95
Sam     |   Science     |   98      |   Language    |   70      |   Social      |   85      |   Maths       |   99      |   English     |   75
Nathan  |   English     |   90      |   Maths       |   99      |   Social      |   85      |   Science     |   85      |   Language    |   90
Hobbs   |   Social      |   85      |   Language    |   90      |   English     |   85      |   Science     |   99      |   Maths       |   85
Shaw    |   Language    |   90      |   Science     |   75      |   Social      |   90      |   Maths       |   95      |   English     |   90

What is done here is that for each name, the subjects are arranged in the order of their occurrence in the original dataframe in columns. What python code can achieve this?

Edit#1: There are other columns like Reg.No, Class, etc. which is just unique value for each student. So I need a solution which has the above mentioned columns along with these.

Answer 1

Try pivot :

In [1415]: df.pivot('Name', 'Subject', 'Mark')
Out[1415]: 
Subject  English  Language  Maths  Science  Social
Name                                              
Daniel        90        95     95       98      99
Hobbs         85        90     85       99      85
Nathan        90        90     99       85      85
Sam           75        70     99       98      85
Shaw          90        90     95       75      90

Answer 2

A little convoluted, but should do:

>>> (df.groupby('Name', sort=False)[['Subject', 'Mark']]
      .apply(lambda x: x.stack().reset_index(drop=True))
      .rename(columns=lambda x: f"{['Subject', 'Mark'][x%2]} {x//2+1}")
    )
       Subject 1  Mark 1 Subject 2  Mark 2 Subject 3  Mark 3 Subject 4  \
Name                                                                     
Daniel     Maths      95    Social      99   English      90   Science   
Sam      Science      98  Language      70    Social      85     Maths   
Nathan   English      90     Maths      99    Social      85   Science   
Hobbs     Social      85  Language      90   English      85   Science   
Shaw    Language      90   Science      75    Social      90     Maths   

        Mark 4 Subject 5  Mark 5  
Name                              
Daniel      98  Language      95  
Sam         99   English      75  
Nathan      85  Language      90  
Hobbs       99     Maths      85  
Shaw        95   English      90   

EDIT:

If you want n subjects:

>>> n = 3
>>> (df.groupby('Name', sort=False)[['Subject', 'Mark']]
      .apply(lambda x: x.iloc[:n].stack().reset_index(drop=True))
      .rename(columns=lambda x: f"{['Subject', 'Mark'][x%2]} {x//2+1}")
    )

       Subject 1  Mark 1 Subject 2  Mark 2 Subject 3  Mark 3
Name                                                        
Daniel     Maths      95    Social      99   English      90
Sam      Science      98  Language      70    Social      85
Nathan   English      90     Maths      99    Social      85
Hobbs     Social      85  Language      90   English      85
Shaw    Language      90   Science      75    Social      90

Answer 3

def flatten(l):
    return [item for sublist in l for item in sublist]

new_df = pd.DataFrame()
index = flatten([["Subject " +str(i), "Mark "+ str(i) ] for i in range(1,6)])
for item in df.groupby('Name'):
    l = (item[1]).iloc[:,1:].values.tolist()
    flat_list = flatten(l)
    s = pd.Series(data = flat_list, index= index, name = item[0])
    new_df = pd.concat([new_df, s], axis=1)
    
new_df = new_df.T