Question

tempdata =pd.pivot_table( gas_diesel_df, values = 'Quantity' ,index = ['Year','Country or Area'],
                      columns = ['Commodity - Transaction'])
tempdata.iloc[0:10]

在这里，我只想过滤特定国家/地区的数量值。但我不能使用loc方法。您对任何数据透视表中的特殊过滤有何建议？

Answer 1

您可以使用xs或loc - docs：

<强>示例

print gas_diesel_df
   Year Country or Area Commodity - Transaction  Quantity
0  2015             A11                      B1        50
1  2015             A11                      C2        60
2  2015             A11                      B1        45
3  2015             A11                      B2        70
4  2015             B11                      B1        40
5  2015             B11                      C2        45
6  2015             C11                      B1        60
7  2015             C11                      B1        65

tempdata =pd.pivot_table( gas_diesel_df, values = 'Quantity' ,index = ['Year','Country or Area'],
                      columns = ['Commodity - Transaction'])
print tempdata
Commodity - Transaction    B1  B2  C2
Year Country or Area                 
2015 A11                 47.5  70  60
     B11                 40.0 NaN  45
     C11                 62.5 NaN NaN

print tempdata.xs('A11', level=1)
Commodity - Transaction    B1  B2  C2
Year                                 
2015                     47.5  70  60
print tempdata.loc[(slice(None),'A11'),:]
Commodity - Transaction    B1  B2  C2
Year                                 
2015                     47.5  70  60

编辑：

如果您需要按多个years选择，请使用loc，请参阅doc：

print gas_diesel_df
   Year Country or Area Commodity - Transaction  Quantity
0  2013             A11                      B1        50
1  2015             A11                      C2        60
2  2015             A11                      B1        45
3  2015             A11                      B2        70
4  2015             B11                      B1        40
5  2014             B11                      C2        45
6  2014             C11                      B1        60
7  2014             C11                      B1        65

tempdata =pd.pivot_table( gas_diesel_df, values = 'Quantity' ,index = ['Year','Country or Area'],
                      columns = ['Commodity - Transaction'])

print tempdata
Commodity - Transaction    B1  B2  C2
Year Country or Area                 
2013 A11                 50.0 NaN NaN
2014 B11                  NaN NaN  45
     C11                 62.5 NaN NaN
2015 A11                 45.0  70  60
     B11                 40.0 NaN NaN

print tempdata.loc[2013]
Commodity - Transaction  B1  B2  C2
Country or Area                    
A11                      50 NaN NaN

print tempdata.loc[2013:2014]
Commodity - Transaction    B1  B2  C2
Year Country or Area                 
2013 A11                 50.0 NaN NaN
2014 B11                  NaN NaN  45
     C11                 62.5 NaN NaN

如何使用数据透视表中的特定索引过滤值（多维数据）

1 个答案: