python pandas将多列索引转换为列值答案

【问题标题】：python pandas turning a multi column index into column valuespython pandas将多列索引转换为列值
【发布时间】：2017-09-01 14:28:28
【问题描述】：

我是 python 新手，正在尝试使用 pandas atm。我知道有很多关于使用索引和列转换数据帧的答案。但是，到目前为止，我还没有找到可以帮助我解决以下问题的答案。我有一个如下所示的数据框：

style                    VALUE       GROWTH        QUALITY
factor          EarningsYield  OPER_MARGIN  RETURN_COM_EQY   GEARING
VEDG LX Equity             NaN    18.604873            NaN  1.04020
DPW DU Equity         0.057845    36.001430      10.957723  0.438649e

我想要实现的是这样的：

ticker           style         factor      value
VEDG LX Equity   VALUE  EarningsYield        NaN
VEDG LX Equity  GROWTH    OPER_MARGIN  18.604873
VEDG LX Equity QUALITY RETURN_COM_EQY        NaN  
VEDG LX Equity QUALITY        GEARING    1.04020
DPW DU Equity    VALUE  EarningsYield   0.057845
DPW DU Equity   GROWTH    OPER_MARGIN  36.001430
DPW DU Equity  QUALITY RETURN_COM_EQY  10.957723
DPW DU Equity  QUALITY        GEARING  0.438649e

任何帮助将不胜感激。请注意，ticker 列当前是索引。

提前谢谢各位。

欢呼

感谢您的回答。到目前为止，我尝试过的如下：

我已经导入了一个 excel 文件，然后这个：

# Load the xls file's Sheet1 as a dataframe
df = xls_file.parse('Sheet1')
style_names = np.array(['VALUE','GROWTH','QUALITY','QUALITY'])
factor_names = np.array([ 'EarningsYield ', 'OPER_MARGIN ' ,'RETURN_COM_EQY','GEARING'])
df.columns = [style_names,factor_names]
df.columns.names = ['style','factor']
print(df)

style                   VALUE       GROWTH        QUALITY          
factor         EarningsYield  OPER_MARGIN  RETURN_COM_EQY   GEARING
VEDG LX Equity            NaN    18.604873            NaN  1.040200
DPW DU Equity        0.057946    36.001430      10.957723  0.438649
SVST LI Equity            NaN    25.405680      41.356272  0.306917
STM IM Equity        0.016426     3.068980       7.371885  0.227296
NYR BB Equity       -0.334866    -5.012305     -32.771536  0.509514
MDC LN Equity        0.000400    13.168425            NaN  0.324293
TIT IM Equity        0.110168    19.563732       6.842755  0.574045
OCI NA Equity       -0.002449    15.971676      12.469365  0.751047
BESI NA Equity       0.031403    20.024775      33.685981  0.263089
IMPN SW Equity       0.041195     2.808368       6.870435  0.390823
MHG NO Equity        0.009682    26.454333      29.083558  0.324450
IAG LN Equity        0.001430    11.450348      42.105263  0.586250
DG FP Equity         0.057341    10.504060      16.673043  0.496945

z = zscore(df)
z[z>3]=3
z[z<-3]=-3

print(z.unstack())

style    factor            
index                    0     VEDG LX Equity
                         1      DPW DU Equity
                         2     SVST LI Equity
                         3      STM IM Equity
                         4      NYR BB Equity
                         5      MDC LN Equity
                         6      TIT IM Equity
                         7      OCI NA Equity
                         8     BESI NA Equity
                         9     IMPN SW Equity
                         10     MHG NO Equity
                         11     IAG LN Equity
                         12      DG FP Equity
VALUE    EarningsYield   0                NaN
                         1           0.509544
                         2                NaN
                         3           0.150815
                         4           -2.88433
                         5          0.0123503
                         6           0.960737
                         7         -0.0122652
                         8           0.280218
                         9           0.364819
                         10         0.0925452
                         11         0.0212482
                         12          0.504316
GROWTH   OPER_MARGIN     0           0.305012
                         1            1.87814
                         2           0.919993
                         3           -1.09986
                                    ...      
                         9           -1.12343
                         10           1.01482
                         11         -0.341955
                         12         -0.427525
QUALITY  RETURN_COM_EQY  0                NaN
                         1          -0.232745
                         2            1.20556
                         3          -0.402409
                         4           -2.30179
                         5                NaN
                         6          -0.427444
                         7          -0.161222
                         8           0.842639
                         9          -0.426135
                         10          0.624876
                         11             1.241
                         12         0.0376744
         GEARING         0            2.49189
                         1           -0.18156
                         2           -0.76701
                         3           -1.12087
                         4           0.133384
                         5          -0.689786
                         6           0.420179
                         7            1.20682
                         8          -0.961795
                         9           -0.39411
                         10          -0.68909
                         11          0.474417
                         12         0.0775232

因此，虽然它看起来更接近我所需要的，但它看起来并不像 Luce 做它时得到的那样。

有什么建议吗？

感谢格里特

更新：

我想我明白了：

df = xls_file.parse('Sheet1')
style_names = np.array(['VALUE','GROWTH','QUALITY','QUALITY'])
factor_names = np.array([ 'EarningsYield ', 'OPER_MARGIN ' ,'RETURN_COM_EQY','GEARING'])

df.columns = [style_names,factor_names]
df.columns.names = ['style','factor']





print(df)


z = zscore(df)
z[z>3]=3
z[z<-3]=-3

print(z)
print(z.unstack())

zu = z.unstack()
zur = zu.reset_index()

print(zur)

现在返回：

   style          factor         level_2         0
0     VALUE  EarningsYield   VEDG LX Equity       NaN
1     VALUE  EarningsYield    DPW DU Equity  0.509544
2     VALUE  EarningsYield   SVST LI Equity       NaN
3     VALUE  EarningsYield    STM IM Equity  0.150815
4     VALUE  EarningsYield    NYR BB Equity -2.884326
5     VALUE  EarningsYield    MDC LN Equity  0.012350
6     VALUE  EarningsYield    TIT IM Equity  0.960737
7     VALUE  EarningsYield    OCI NA Equity -0.012265
8     VALUE  EarningsYield   BESI NA Equity  0.280218
9     VALUE  EarningsYield   IMPN SW Equity  0.364819
10    VALUE  EarningsYield    MHG NO Equity  0.092545
11    VALUE  EarningsYield    IAG LN Equity  0.021248
12    VALUE  EarningsYield     DG FP Equity  0.504316
13   GROWTH    OPER_MARGIN   VEDG LX Equity  0.305012
14   GROWTH    OPER_MARGIN    DPW DU Equity  1.878142
15   GROWTH    OPER_MARGIN   SVST LI Equity  0.919993
16   GROWTH    OPER_MARGIN    STM IM Equity -1.099862
17   GROWTH    OPER_MARGIN    NYR BB Equity -1.830634
18   GROWTH    OPER_MARGIN    MDC LN Equity -0.186593
19   GROWTH    OPER_MARGIN    TIT IM Equity  0.391720
20   GROWTH    OPER_MARGIN    OCI NA Equity  0.066898
21   GROWTH    OPER_MARGIN   BESI NA Equity  0.433411
22   GROWTH    OPER_MARGIN   IMPN SW Equity -1.123429
23   GROWTH    OPER_MARGIN    MHG NO Equity  1.014821
24   GROWTH    OPER_MARGIN    IAG LN Equity -0.341955
25   GROWTH    OPER_MARGIN     DG FP Equity -0.427525
26  QUALITY  RETURN_COM_EQY  VEDG LX Equity       NaN
27  QUALITY  RETURN_COM_EQY   DPW DU Equity -0.232745
28  QUALITY  RETURN_COM_EQY  SVST LI Equity  1.205558
29  QUALITY  RETURN_COM_EQY   STM IM Equity -0.402409
30  QUALITY  RETURN_COM_EQY   NYR BB Equity -2.301789
31  QUALITY  RETURN_COM_EQY   MDC LN Equity       NaN
32  QUALITY  RETURN_COM_EQY   TIT IM Equity -0.427444
33  QUALITY  RETURN_COM_EQY   OCI NA Equity -0.161222
34  QUALITY  RETURN_COM_EQY  BESI NA Equity  0.842639
35  QUALITY  RETURN_COM_EQY  IMPN SW Equity -0.426135
36  QUALITY  RETURN_COM_EQY   MHG NO Equity  0.624876
37  QUALITY  RETURN_COM_EQY   IAG LN Equity  1.240996
38  QUALITY  RETURN_COM_EQY    DG FP Equity  0.037674
39  QUALITY         GEARING  VEDG LX Equity  2.491893
40  QUALITY         GEARING   DPW DU Equity -0.181560
41  QUALITY         GEARING  SVST LI Equity -0.767010
42  QUALITY         GEARING   STM IM Equity -1.120867
43  QUALITY         GEARING   NYR BB Equity  0.133384
44  QUALITY         GEARING   MDC LN Equity -0.689786
45  QUALITY         GEARING   TIT IM Equity  0.420179
46  QUALITY         GEARING   OCI NA Equity  1.206820
47  QUALITY         GEARING  BESI NA Equity -0.961795
48  QUALITY         GEARING  IMPN SW Equity -0.394110
49  QUALITY         GEARING   MHG NO Equity -0.689090
50  QUALITY         GEARING   IAG LN Equity  0.474417
51  QUALITY         GEARING    DG FP Equity  0.077523

感谢您为我指明正确的方向！

非常感谢

【问题讨论】：

到目前为止你尝试过什么？请展示一些努力/代码。

标签： python pandas transform multi-index

【解决方案1】：

您似乎正在尝试转置矩阵。这是使用 pandas 的一种方法：

import pandas as pd
from io import StringIO
csv_data = '''A,B,C,D
1.0, 2.0, 3.0, 4.0
5.0, 6.0, 7.0, 8.0
9.0, 10.0, 11.0, 12.0'''
df = pd.read_csv(StringIO(csv_data))
df.transpose()

【讨论】：

【解决方案2】：

Unstack可用于将宽格式转长格式（相反的功能是pivot）。

>> import pandas as pd
>> from io import StringIO

>> csv = StringIO(u'''style VALUE   GROWTH  QUALITY QUALITY
factor  EarningsYield   OPER_MARGIN RETURN_COM_EQY  GEARING
VEDG LX Equity  NaN 18.604873   NaN 1.04020
DPW DU Equity   0.057845    36.001430   10.957723   0.438649e''')

>> df = pd.read_csv(csv, sep='\t', header=[0, 1], index_col=0)
>> df
style                  VALUE      GROWTH        QUALITY           
factor         EarningsYield OPER_MARGIN RETURN_COM_EQY    GEARING
VEDG LX Equity           NaN   18.604873            NaN    1.04020
DPW DU Equity       0.057845   36.001430      10.957723  0.438649e

>> df.unstack()
style    factor                        
VALUE    EarningsYield   VEDG LX Equity          NaN
                         DPW DU Equity      0.057845
GROWTH   OPER_MARGIN     VEDG LX Equity      18.6049
                         DPW DU Equity       36.0014
QUALITY  RETURN_COM_EQY  VEDG LX Equity          NaN
                         DPW DU Equity       10.9577
         GEARING         VEDG LX Equity      1.04020
                         DPW DU Equity     0.438649e

要获得确切的格式，您必须尝试一下。

【讨论】：