Filter dataframe on second row using list

Question

I have a dataframe that needs to be kept in the structure below.

df
  One       two   three          
  date      apple  banana
2019-12-20   0      4
2020-01-03   10     5

Is there a way to keep only columns where the second row is contained within the list? I'm only familiar with doing this for columns

fruits= ["date", "apple"]
df = df.loc[:, df.columns.isin(fruits)

Expected Output

   One       two          
  date      apple  
2019-12-20   0      
2020-01-03   10    

Answer 1

use .isin and use the level argument to specify the level of the multiindex.

if you don't have a multiindex set, you can set it on your read.

df = pd.read_csv('file.csv',sep='\s+',header=[0,1])

df.loc[:,df.columns.isin(fruits,level=1)]

print(df)
         One   two
         date apple
2  2019-12-20     0
3  2020-01-03    10

---

df.columns.isin(fruits,level=1)
array([ True,  True, False])