Pandas 通过保留一列的第一个元素和另一列的最后一个元素来合并

Question

I have a dataframe with different value and ID that can be in common.

    df = pd.DataFrame({'A': ['chr1','chr1','chr1','chr1','chr1','chr2'],
                    'B': [700,750,800,850,900,200],
                    'C': [750,800,850,900,950,250],
                    'D':['id_1','id_1','id_1','id_1','id_1','id_2']})

What I'm trying to do is to keep lowest element of B, higher of C for identical value A and D

Output should look :

    A    B    C    D
0  chr1 700  950   id_1
1  chr2 200  250   id_2

I tried to use

groupby('D').agg(['first', 'last'])

But it's not what I want...

Answer 1

Use GroupBy.agg with dictionary by columns names and aggregate functions:

df1 = (df.groupby('D', as_index=False)
         .agg({'A':'first', 'B':'first', 'C':'last'})
         [['A','B','C','D']])
print (df1)
      A    B    C     D
0  chr1  700  950  id_1
1  chr2  200  250  id_2

Answer 2

With dict passed the name and function inside the agg

df.groupby(['A','D'],as_index=False).agg({'B':'first','C':'last'}).reindex(columns=df.columns)
      A    B    C     D
0  chr1  700  950  id_1
1  chr2  200  250  id_2