Reshape DataFrame Pandas - 一些变量变长，另一些变量变宽

Question

我需要重塑数据框，以便某些变量（Diag1、Diag2、Diag3）变长，而其他变量（句点）变宽。基本上他们需要交换位置。

我在下面的示例中重新创建了原始数据框。如示例所示，我已尝试分别使用 pivot 和 melt 无济于事。

df = pd.DataFrame({
    'ID':[1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,],
    'Period':['0 Month','3 Month','0 Month','3 Month','0 Month',
            '3 Month','0 Month','3 Month','0 Month','3 Month','0 Month',
            '3 Month','0 Month','3 Month','0 Month','3 Month','0 Month',
            '3 Month','0 Month','3 Month','0 Month','3 Month',],
    'Diag1':[0,1,1,0,0,0,0,1,0,0,1,0,0,0,0,0,0,0,1,0,0,0,],
    'Diag2':[0,0,1,0,0,0,0,1,0,0,1,0,0,0,0,0,0,0,1,0,0,0,],
    'Diag3':[0,0,1,0,0,0,1,1,0,0,1,0,0,0,1,0,0,0,1,0,1,0,]
    })

dfp = df.pivot(index=["ID",], columns='Period',).reset_index()
print(dfp)

dfm = df.melt(id_vars=["ID",],value_vars=['Period'])
print(dfm)

期望的结果是：

ID  Diagnosis   0_Month 3_Month
1   Diag1       0       1
1   Diag2       0       0
1   Diag3       0       0
2   Diag1       0       1
2   Diag2       1       0
2   Diag3       1       0
3   Diag1   

3   Diag2   

3   Diag3       etc...

我怀疑我需要两者的某种组合，但正在努力寻找任何示例。结果我的大脑开始融化了...

Answer 1

你可以melt；然后 pivot:

out = (df.melt(id_vars=['ID', 'Period'], var_name='Diagnosis')
       .pivot(['ID','Diagnosis'], 'Period', 'value')
       .reset_index().rename_axis(columns=[None]))

输出：

    ID Diagnosis  0 Month  3 Month
0    1     Diag1        0        1
1    1     Diag2        0        0
2    1     Diag3        0        0
3    2     Diag1        1        0
4    2     Diag2        1        0
..  ..       ...      ...      ...
28  10     Diag2        1        0
29  10     Diag3        1        0
30  11     Diag1        0        0
31  11     Diag2        0        0
32  11     Diag3        1        0

[33 rows x 4 columns]

Reshape DataFrame Pandas - 一些变量变长，另一些变量变宽

Reshape DataFrame Pandas - some variables to long others to wide

python

pandas

dataframe

pivot-table

pandas-melt