python - 值错误 : Grouper for not 1-dimensional

Question

我有以下代码通过 seaborn 创建一个表格和一个条形图。

#Building a dataframe grouped by the # of Engagement Types
sales_type = sales.groupby('# of Engagement Types').sum()

#Calculating the % of people who bought the course by # engagement types
sales_type['% Sales per Participants'] =  round(100*(sales_type['Sales'] / sales_type['Had an Engagement']), 2)

#Calculating the # of people who didn't have any engagements
sales_type.set_value(index=0, col='Had an Engagement', value=sales[sales['Had an Engagement']==0].count()['Sales'])

#Calculating the % of sales for those who didn't have any engagements
sales_type.set_value(index=0, col='% Sales per Participants',
                     value=round(100 * (sales_type.ix[0, 'Sales'] / 
                                        sales[sales['Had an Engagement']==0].count()['Sales']),2))

#Setting the graph image
fig, (ax1) = plt.subplots(nrows=1, ncols=1, figsize=(12,4))
sns.set_style("whitegrid")

# Ploting the histagram for the % of total prospects
ax1 = sns.barplot(x=sales_type.index,y='% Sales per Participants', data=sales_type ,ax=ax1)
ax1.set(ylabel = '%')
ax1.set_title('% Sales per Participants By # of Engagement Types') 

#present the table
sales_type.xs(['Had an Engagement', 'Sales','% Sales per Participants'],axis=1).transpose()
#sales_type

我对其他没有问题的参数使用相同的代码概念。但是，对于一个参数，我收到一个错误:行代码的“ValueError: Grouper for '' not 1-dimensional”:

ax1 = sns.barplot(x=sales_type.index,y='% Sales per Participants', data=sales_type ,ax=ax1)

尽管数据框的维度不超过一维，但仍会发生此错误。

这是 table 的头:

                       Sales  Pre-Ordered / Ordered Book  \
# of Engagement Types                                      
0                        1.0                         0.0   
1                       20.0                       496.0   
2                       51.0                       434.0   
3                       82.0                       248.0   
4                       71.0                       153.0   
5                       49.0                        97.0   
6                        5.0                        24.0   

                       Opted In For / Clicked to Kindle  Viewed PLC  \
# of Engagement Types                                                 
0                                                   0.0           0   
1                                               27034.0        5920   
2                                                6953.0        6022   
3                                                1990.0        1958   
4                                                 714.0         746   
5                                                 196.0         204   
6                                                  24.0          24   

                       # of PLC Engagement  Viewed Webinar  \
# of Engagement Types                                        
0                                      0.0               0   
1                                   6434.0            1484   
2                                   7469.0            1521   
3                                   2940.0            1450   
4                                   1381.0             724   
5                                    463.0             198   
6                                     54.0              24   

                       # of Webinars (Live/Replay)  \
# of Engagement Types                                
0                                              0.0   
1                                           1613.0   
2                                           1730.0   
3                                           1768.0   
4                                           1018.0   
5                                            355.0   
6                                             45.0   

                       OCCC Facebook Group Member  Engaged in Cart-Open  \
# of Engagement Types                                                     
0                                             0.0                     0   
1                                           148.0                   160   
2                                           498.0                  1206   
3                                           443.0                   967   
4                                           356.0                   511   
5                                           168.0                   177   
6                                            24.0                    24   

                       # of Engagement at Cart Open  Had an Engagement  \
# of Engagement Types                                                    
0                                               0.0               3387   
1                                             189.0              35242   
2                                            1398.0               8317   
3                                            1192.0               2352   
4                                             735.0                801   
5                                             269.0                208   
6                                              40.0                 24   

                       Total # of Engagements  % Sales per Participants  
# of Engagement Types                                                    
0                                         0.0                      0.03  
1                                     35914.0                      0.06  
2                                     18482.0                      0.61  
3                                      8581.0                      3.49  
4                                      4357.0                      8.86  
5                                      1548.0                     23.56  
6                                       211.0                     20.83  

这是完整的错误:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-211-f0185fe64c1a> in <module>()
     12 sns.set_style("whitegrid")
     13 # Ploting the histagram for the % of total prospects
---> 14 ax1 = sns.barplot(x=sales_type.index,y='% Sales per Participants', data=sales_type ,ax=ax1)
     15 ax1.set(ylabel = '%')
     16 ax1.set_title('% Sales per Participants By # of Engagement Types')

ValueError: Grouper for '<class 'pandas.core.frame.DataFrame'>' not 1-dimensional

我已尝试在 Internet 和 Stack Overflow 上搜索此错误，但没有得到任何结果。有人知道发生了什么吗？

Answer 1

简化问题

我也遇到了这个问题，发现是列名重复造成的。

要重新创建这个:

df = pd.DataFrame({"foo": [1,2,3], "bar": [1,2,3]})
df.rename(columns={'foo': 'bar'}, inplace=True)

   bar  bar
0    1    1
1    2    2
2    3    3

df.groupby('bar')

ValueError: Grouper for 'bar' not 1-dimensional

就像很多神秘的 pandas 错误一样，这个错误也源于有两个同名的列。

确定您要使用哪一列，重命名或删除另一列并重做操作。

解决方案

像这样重命名列

df.columns = ['foo', 'bar']

   foo  bar
0    1    1
1    2    2
2    3    3

df.groupby('bar')
<pandas.core.groupby.DataFrameGroupBy object at 0x1066dd950>