必须重新检查列名。Days与days

还要检查列名称是否由于某种原因而丢失。可能是合并的结果
df.columns

试试

df.drop_duplicates(subset=['id', 'event_description', 'attribute1', 'days', 'days_supply', 'days_equivalent'],inplace=True)

出发地：https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.drop_duplicates.html
试试
也许你的df格式不好，无论如何如果你认为这个问题与dtype有关，你可以使用函数apply来检查df ['date']的整个数据，如下所示：

def checkType(someDate):
    ##Do verification
    return dateCorrected

df['date'] = df['date'].apply(checkType)

我重现了一个有点类似的情况：列配置错误（一对多余的方括号）的DataFrame返回一个看起来不错的结果（图1）。

array = [
    ['001', 3, 3, 3, 1, 5, 4, 3],
    ['002', 7, 2, 1, 1, 1, 5, 1],
    ['003', 1, 6, 7, 6, 6, 7, 7]]

# NG configuration of the columns.
df_NG = pd.DataFrame(
    array,
    columns=[
        ['id', 'event_description', 'attribute1', 'attribute 2', 'attribute 3',
         'days', 'days_supply', 'days_equivalent']])

图1伪OK数据框（内部腐烂）x1c 0d1x
但如果你想删除重复的，

df_NG = df_NG.drop_duplicates(
    subset=[
        'id', 'event_description', 'attribute1',
        'days', 'days_supply', 'days_equivalent'])

Pandas归来：

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
Input In [71], in <cell line: 1>()
----> 1 df_NG = df_NG.drop_duplicates(
      2     subset=[
      3         'id', 'event_description', 'attribute1',
      4         'days', 'days_supply', 'days_equivalent'])

File /usr/local/lib/python3.9/site-packages/pandas/util/_decorators.py:311, in deprecate_nonkeyword_arguments.<locals>.decorate.<locals>.wrapper(*args, **kwargs)
    305 if len(args) > num_allow_args:
    306     warnings.warn(
    307         msg.format(arguments=arguments),
    308         FutureWarning,
    309         stacklevel=stacklevel,
    310     )
--> 311 return func(*args, **kwargs)

File /usr/local/lib/python3.9/site-packages/pandas/core/frame.py:6125, in DataFrame.drop_duplicates(self, subset, keep, inplace, ignore_index)
   6123 inplace = validate_bool_kwarg(inplace, "inplace")
   6124 ignore_index = validate_bool_kwarg(ignore_index, "ignore_index")
-> 6125 duplicated = self.duplicated(subset, keep=keep)
   6127 result = self[-duplicated]
   6128 if ignore_index:

File /usr/local/lib/python3.9/site-packages/pandas/core/frame.py:6259, in DataFrame.duplicated(self, subset, keep)
   6257 diff = Index(subset).difference(self.columns)
   6258 if not diff.empty:
-> 6259     raise KeyError(diff)
   6261 vals = (col.values for name, col in self.items() if name in subset)
   6262 labels, shape = map(list, zip(*map(f, vals)))

KeyError: Index(['attribute1', 'days', 'days_equivalent', 'days_supply',
       'event_description', 'id'],
      dtype='object')

所以我跟踪David's suggestion找到了罪魁祸首！

>>> df_NG.columns

MultiIndex([(               'id',),
        ('event_description',),
        (       'attribute1',),
        (      'attribute 2',),
        (      'attribute 3',),
        (             'days',),
        (      'days_supply',),
        (  'days_equivalent',)],
       )

当然，正确的配置如下所示：）

df_OK = pd.DataFrame(
    array,
    columns=[
        'id', 'event_description', 'attribute1', 'attribute 2', 'attribute 3',
        'days', 'days_supply', 'days_equivalent'])