Question

当我想将数字转换为分类时，以下代码不起作用

data_train['review_scores_rating'] = pd.DataFrame({'value': np.random.randint(20, 100, 1000)}) 
data_train['review_scores_rating'] = pd.cut(data_train['review_scores_rating'].value,
                     bins=[0, 50,80, np.inf],
                     labels=["Detractors", "Passives", "Promoters"])

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-49-ba9263d7208f> in <module>
      1 data_train['review_scores_rating'] = pd.DataFrame({'value': np.random.randint(20, 100, 1000)})
----> 2 data_train['review_scores_rating'] = pd.cut(data_train['review_scores_rating'].value,
      3                      bins=[0, 50,80, np.inf],
      4                      labels=["Detractors", "Passives", "Promoters"])

/anaconda3/lib/python3.7/site-packages/pandas/core/generic.py in __getattr__(self, name)
   5065             if self._info_axis._can_hold_identifiers_and_holds_name(name):
   5066                 return self[name]
-> 5067             return object.__getattribute__(self, name)
   5068 
   5069     def __setattr__(self, name, value):

AttributeError: 'Series' object has no attribute 'value'

Answer 1

我相信这一行：

data_train['review_scores_rating'] = pd.DataFrame({'value': np.random.randint(20, 100, 1000)})

没有做你想做的事。

print(type(pd.DataFrame({'value': np.random.randint(20, 100, 1000)})))
> output: <class 'pandas.core.frame.DataFrame'>

print(type(data_train['review_scores_rating']))
> output: <class 'pandas.core.series.Series'>

熊猫似乎将分配给review_scores_rating的DataFrame转换为Series，这意味着列value不再存在。我对熊猫还不够熟悉，无法说明为什么会这样，但是我认为不能将DataFrame分配给列是有道理的。

您提供的代码的预期输出是什么？您是否要生成多维DataFrame？

如何解决“ AttributeError：'系列'对象没有属性'值'”

1 个答案: