CODEWITHSUNDEEP
pythonpython-3.xpandasdataframesorting
Mayank Porwal
Mayank Porwal

Reputation: 34086

Pandas: Sort innermost column group-wise based on other multilevel column excluding one row

This is an extension to my previous question:

Consider below df:

In [68]: df = pd.DataFrame({'A': ['a'] * 11,
    ...:                'B': ['b'] * 11,
    ...:                'C': ['C1', 'C1', 'C2','C1', 'C3', 'C3', 'C2', 'C3', 'C3', 'C2', 'C2'],
    ...:                'D': ['D1', 'D2', 'D1', 'D3', 'D3', 'D2', 'D4', 'D4', 'D1', 'D2', 'D3'],
    ...:                'E': [{'value': '4', 'percentage': None}, {'value': 5, 'percentage': None}, {'value': 12, 'percentage': None}, {'value': 9, 'percentage': None}, {'value': '12', 'percentage': None}, {'value': 'N/A', 'percentage': None}, {}, {'valu
    ...: e': 24, 'percentage': None}, {'value': 12, 'percentage': None}, {'value': 33, 'percentage': None}, {'value': 11, 'percentage': None}]})
    ...:

Pivot of above df:

In [69]: x = df.pivot(['B', 'C', 'D'], 'A', ['E'])

In [70]: x
Out[70]: 
                                            E
A                                           a
B C  D                                       
b C1 D1    {'value': '4', 'percentage': None}
     D2      {'value': 5, 'percentage': None}
     D3      {'value': 9, 'percentage': None}
  C2 D1     {'value': 12, 'percentage': None}
     D2     {'value': 33, 'percentage': None}
     D3     {'value': 11, 'percentage': None}
     D4                                    {}
  C3 D1     {'value': 12, 'percentage': None}
     D2  {'value': 'N/A', 'percentage': None}
     D3   {'value': '12', 'percentage': None}
     D4     {'value': 24, 'percentage': None}

I want to sort the innermost column which is D for each group of outer columns B and C based on the multi-level column with index (E, a) in asc/desc order based on value key from dict.

But, for every group there would be a row with the total value of all other rows. I always need to keep that row at the last irrespective of sorting order(asc or desc).

Expected output in case of desc:

Out[70]: 
                                            E
A                                           a
B C  D                                       
b C1 D2      {'value': 5, 'percentage': None}
     D1    {'value': '4', 'percentage': None}
     D3      {'value': 9, 'percentage': None}
  C2 D1     {'value': 12, 'percentage': None}
     D3     {'value': 11, 'percentage': None}
     D4                                    {}
     D2     {'value': 33, 'percentage': None}
  C3 D1     {'value': 12, 'percentage': None}
     D3   {'value': '12', 'percentage': None}
     D2  {'value': 'N/A', 'percentage': None}
     D4     {'value': 24, 'percentage': None}

Expected output in case of asc:

Out[70]: 
                                            E
A                                           a
B C  D                                       
b C1 D1    {'value': '4', 'percentage': None}
     D2      {'value': 5, 'percentage': None}
     D3      {'value': 9, 'percentage': None}
  C2 D3     {'value': 11, 'percentage': None}
     D1     {'value': 12, 'percentage': None}
     D4                                    {}
     D2     {'value': 33, 'percentage': None}
  C3 D1     {'value': 12, 'percentage': None}
     D3   {'value': '12', 'percentage': None}
     D2  {'value': 'N/A', 'percentage': None}
     D4     {'value': 24, 'percentage': None}

Upvotes: 2

Views: 90

Answers (2)

jezrael
jezrael

Reputation: 863236

Solution with helper columns used for sorting - first convert values to numeric columns by Series.str.get and to_numeric and then create another column compared max value per groups for boolean:

lvls = list(x.index.names[:-1])
print (lvls)
['B', 'C']

x[('tmp', 'tmp')] = pd.to_numeric(x[('E','a')].str.get('value'), errors='coerce')

x[('max','tmp')] = x.groupby(lvls)[[('tmp','tmp')]].transform('max') == x[[('tmp','tmp')]]

All values in ascending parameter are True, default value:

x1 = x.sort_values(lvls + [('max','tmp'), ('tmp','tmp')])
print (x1)
                                            E   tmp    max
A                                           a   tmp    tmp
B C  D                                                    
b C1 D1    {'value': '4', 'percentage': None}   4.0  False
     D2      {'value': 5, 'percentage': None}   5.0  False
     D3      {'value': 9, 'percentage': None}   9.0   True
  C2 D3     {'value': 11, 'percentage': None}  11.0  False
     D1     {'value': 12, 'percentage': None}  12.0  False
     D4                                    {}   NaN  False
     D2     {'value': 33, 'percentage': None}  33.0   True
  C3 D1     {'value': 12, 'percentage': None}  12.0  False
     D3   {'value': '12', 'percentage': None}  12.0  False
     D2  {'value': 'N/A', 'percentage': None}   NaN  False
     D4     {'value': 24, 'percentage': None}  24.0   True

Here is changed last True to False:

x2 = x.sort_values(lvls + [('max','tmp'), ('tmp','tmp')],
                   ascending=[True] * len(lvls) + [True, False])
print (x2)

                                            E   tmp    max
A                                           a   tmp    tmp
B C  D                                                    
b C1 D2      {'value': 5, 'percentage': None}   5.0  False
     D1    {'value': '4', 'percentage': None}   4.0  False
     D3      {'value': 9, 'percentage': None}   9.0   True
  C2 D1     {'value': 12, 'percentage': None}  12.0  False
     D3     {'value': 11, 'percentage': None}  11.0  False
     D4                                    {}   NaN  False
     D2     {'value': 33, 'percentage': None}  33.0   True
  C3 D1     {'value': 12, 'percentage': None}  12.0  False
     D3   {'value': '12', 'percentage': None}  12.0  False
     D2  {'value': 'N/A', 'percentage': None}   NaN  False
     D4     {'value': 24, 'percentage': None}  24.0   True

Last remove helper columns:

x1 = x1.drop([('max','tmp'), ('tmp','tmp')], axis=1)
x2 = x2.drop([('max','tmp'), ('tmp','tmp')], axis=1)

Upvotes: 2

Shubham Sharma
Shubham Sharma

Reputation: 71687

You can define a function that groups the multilevel column E, a on levels B and C and returns the indices that would sort the dataframe as specified by the rule where the row with the total value of all other rows is kept at last:

def sort_idx(s):
    idx = []
    for k, g in s.groupby(level=[0, 1], sort=False):
        i = g.idxmax()
        idx += [*g.drop(i).sort_values().index , i]
    return idx

s = pd.to_numeric(x[('E', 'a')].str['value'], errors='coerce')
x = x.loc[sort_idx(s)]

Result:

                                            E
A                                           a
B C  D                                       
b C1 D1    {'value': '4', 'percentage': None}
     D2      {'value': 5, 'percentage': None}
     D3      {'value': 9, 'percentage': None}
  C2 D3     {'value': 11, 'percentage': None}
     D1     {'value': 12, 'percentage': None}
     D4                                    {}
     D2     {'value': 33, 'percentage': None}
  C3 D1     {'value': 12, 'percentage': None}
     D3   {'value': '12', 'percentage': None}
     D2  {'value': 'N/A', 'percentage': None}
     D4     {'value': 24, 'percentage': None}

Upvotes: 2

Related Questions