复制列时出现意外行为

Question

我有一个包含多个列的数据框（df），其中一列是loan_status。该栏的细目如下：

AnyOneEvenFromUnsafeGroups

我想创建一个新列，将每个条目分类为具有良好信誉或不良信誉的贷款。我做了以下事情：

Charged Off                                              38029
Current                                                 520784
Default                                                    199
Does not meet the credit policy.  Status:Charged Off       758
Does not meet the credit policy.  Status:Current            15
Does not meet the credit policy.  Status:Fully Paid       1947
Fully Paid                                              176203
In Grace Period                                           6486
Issued                                                     142
Late (16-30 days)                                         2337
Late (31-120 days)                                        9949

这给了我以下错误：

  

ValueError：无法使用包含NA / NaN值的向量进行索引

这让我意识到数据中必须有一些纳米值，而且出于我的好奇心：

df["badloan"] = df.loan_status
df.loc[(df.badloan == "Charged Off") | 
       (df.badloan == "Default") |      
       (df.badloan =="Does not meet the credit policy.  Status:Charged Off"),
       df.badloan] = "Bad"

这导致了大多数真正的价值，但有些是假的。然后我尝试重新定义df.badloan，以防我意外修改它：

df["badloan"] == df.loan_status

这仍然产生了一些错误的价值。知道为什么吗？