解决Pandas中按参赛条件删DataFrame行的ValueError问题

问题场景 #

需要删除ratings数据框中未参赛球队的行（球队未出现在fixtures的HomeTeam或AwayTeam列），原代码运行报错：

fixtures = pd.DataFrame({'HomeTeam': ["Team1", "Team3", "Team5", "Team6"], 'AwayTeam': [
    "Team2", "Team4", "Team6", "Team8"]})

ratings = pd.DataFrame({'team': ["Team1", "Team2", "Team3", "Team4", "Team5",
                                 "Team6", "Team7", "Team8", "Team9", "Team10", "Team11", "Team12"], "rating": ["1,5", "0,2", "0,5", "2", "3", "4,8", "0,9", "-0,4", "-0,6", "1,5", "0,2", "0,5"]})

ratings = ratings[(ratings.team != fixtures.HomeTeam) &
                  (ratings.team != fixtures.AwayTeam)]

错误信息：

ValueError: Can only compare identically-labeled Series objects

错误原因 #

直接用ratings.team != fixtures.HomeTeam这类方式比较两个Series时，Pandas要求两个Series必须标签完全匹配、长度一致。但这里ratings.team长度为12，fixtures.HomeTeam长度为4，无法完成逐元素的匹配对比，因此触发ValueError。

正确解决方法 #

先收集所有参赛球队的集合，再用isin()方法筛选符合条件的行：

# 合并主场和客场球队，去重得到所有参赛球队
participating_teams = pd.concat([fixtures.HomeTeam, fixtures.AwayTeam]).unique()

# 保留ratings中属于参赛球队的行（删除未参赛球队）
ratings = ratings[ratings.team.isin(participating_teams)]

执行后，ratings将仅保留Team1至Team8的记录，Team9至Team12这些未参赛球队会被移除。

如果需要直接筛选未参赛球队（即删除参赛的），可以用~取反：

ratings = ratings[~ratings.team.isin(participating_teams)]

内容的提问来源于stack exchange，提问作者Robin Reiche