使用 python 从 df 中删除不在另一个 df 中的记录

Question

I've a sample datafram1

date           username         cities
2021-03-01     K John           New york
2021-03-01     K John           LA
2021-03-02     Ken Miles        Florida
2021-03-02     Ken Miles        LA

dataframe2 contains

date          username        planned_cities 
2021-03-01    K John             Alabama
2021-03-02    K John             LA
2021-03-02    Ken Miles          Florida
2021-03-02    Ken Miles          California

Expected result (by considering only date username , dropping the columns that are not in df1)

date         username        planned_cities
2021-03-01    K John             Alabama
2021-03-02    Ken Miles          Florida
2021-03-02    Ken Miles          California

As 2021-03-02 K John is not in records of df1, it is dropped. How could I achieve this?

Answer 1

you could use Index.isin with the columns you are interested in and then boolean index:

cols = ['date','username']
idx1 = pd.MultiIndex.from_frame(df1[cols])
idx2 = pd.MultiIndex.from_frame(df2[cols])
out = df2[idx2.isin(idx1)]

---

       date   username planned_cities
  2021-03-01     K John        Alabama
  2021-03-02  Ken Miles        Florida
  2021-03-02  Ken Miles     California

Answer 2

Use an inner merge dropping duplicates that way you ensure you don't grow the left DataFrame.

df2.merge(df1[['date', 'username']].drop_duplicates())

         date   username planned_cities
0  2021-03-01     K John        Alabama
1  2021-03-02  Ken Miles        Florida
2  2021-03-02  Ken Miles     California