如果我听得懂你的意思那你就想这么做

df1 = pd.DataFrame({'PK': [11, 13, 15, 17, 20]})
df2 = pd.DataFrame({'PK': [11, 12, 13, 15, 16, 17, 20, 25]})
to_remove = set(df2['PK']) - set(df1['PK'])
df_remove = pd.DataFrame({'PK': list(to_remove)})

# Merge df_remove with df2 using an inner join
df2_new = df2.merge(df_remove, on='PK', how='inner')

我相信你想要实现的是以下几点：

df1 = pd.DataFrame({'PK': [11, 13, 15, 17, 20]})
df2 = pd.DataFrame({'PK': [11, 12, 13, 15, 16, 17, 20, 25]})
df = df2[df2['PK'].isin(df1['PK'])]

# if you want the opposite then you can do the following:
df = df2[~df2['PK'].isin(df1['PK'])]

我想你在找这样的东西

import pandas as pd
# creating the dataframes:

df1 = pd.DataFrame({'ENV1': [11,13,15,17,20]})
df2 = pd.DataFrame({'ENV2': [11,12,13,15,16,17,20,25]})


df3 = (df1.merge(df2, right_on='ENV2', left_on='ENV1', how='right', indicator=True)
     .query('_merge == "right_only"')
     .drop('_merge', 1))

df3

然后从df2中删除（12，16，25）：

df2 = df2.drop(df3.index, axis=0) # Step 2
df2