python-3.x - 在尝试执行复杂查询（Pandas）时收到错误“无法使用 dtyped [object] 数组和 [bool] 类型的标量执行 'ror_'”

Question

我有一个熊猫数据框，例如：

CLIENT | CAMPAING START DATE
CLIENT1| Jul/21
CLIENT2| Aug/21

我想删除某个客户和某个月份的一些行。我写：

df.loc[~((df['CLIENT'].str.contains('CLIENT1') & (df['CAMPAIGN START DATE']=="Jul/21" | df['CAMPAIGN START DATE']=="Aug/21" | df['CAMPAIGN START DATE']=="Sep/21" | df['CAMPAIGN START DATE']=="Oct/21" | df['CAMPAIGN START DATE']=="Nov/21" | df['CAMPAIGN START DATE']=="Dec/21") ))]

我收到此错误消息：

TypeError: Cannot perform 'ror_' with a dtyped [object] array and scalar of type [bool]

为什么？

标签： python-3.xpandas

Answer 1

在您的解决方案中缺少()，但可以通过以下方式简化测试成员的解决方案Series.isin：

df.loc[~((df['CLIENT'].str.contains('CLIENT1') & ((df['CAMPAIGN START DATE']=="Jul/21") | (df['CAMPAIGN START DATE']=="Aug/21") | (df['CAMPAIGN START DATE']=="Sep/21") | (df['CAMPAIGN START DATE']=="Oct/21" )| (df['CAMPAIGN START DATE']=="Nov/21") | (df['CAMPAIGN START DATE']=="Dec/21")) ))]

---

df.loc[~(df['CLIENT'].str.contains('CLIENT1') & 
         df['CAMPAIGN START DATE'].isin(["Jul/21","Aug/21","Sep/21","Oct/21","Nov/21","Dec/21"])) ]