熊猫删除每个组中每个元素在列上具有相同值的组

Question

我需要为df 做一个groupby，然后在每个组中，我想检查该组中的每个元素在A 列上是否具有相同的值，如果是，则删除该组,

 df['cluster_id'] = df.groupby(['B', 'C', 'D'])['B'].transform('size')

 df = df.loc[
        df['cluster_id'] > 1 &
        df['cluster_id'] == df['cluster_id'] &
        df['A'] != df['A']]

但我有错误

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

我想知道如何解决它。

Answer 1

我猜() 不见了：

df =df[(df['cluster_id'] > 1) & (df['cluster_id'] == df['cluster_id']) & (df['A'] != df['A'])]

另外，似乎第二个条件是不必要的：

df = df[(df['cluster_id'] > 1) & (df['A'] != df['A'])]

也不需要新列，可以通过Series进行比较：

cluster_id = df.groupby(['B', 'C', 'D'])['B'].transform('size')

df = df[(cluster_id > 1) & (cluster_id == cluster_id) & (df['A'] != df['A'])]

---

df = df[(cluster_id > 1) & (df['A'] != df['A'])]