# define a list of scores l = [10,18,20] # filter out records by scores by list l records = df.filter(~df.score.isin(l))