Python 在dataframe上使用apply（），并将其他dataframe列作为输入

我有一个足球比赛结果的数据框，我正试图在数据框的末尾创建一个新的列，显示哪支球队获胜。我正试图使用df.apply来实现这一点。以下是我到目前为止的情况：

def match_winner(winner,home_team,away_team,home_goals,away_goals):

    if home_goals>away_goals:
        winner = home_team
    elif home_goals<away_goals:
        winner = away_team
    else:
        winner = "None"

match['Winning Team'] =""
match['Winning Team'].apply(match_winner,args=[match['Home Team'],match['Away Team'],match['home_team_goal'],match['away_team_goal']])

我需要发送多个列作为输入参数，以便找到谁赢了，但我不确定.apply是否允许您将一系列作为参数传递？是否可以使用.apply执行此操作。如果是这样的话，一个解决方案会很有帮助，但是如果有人知道一个更好的选择，那也是有用的。

< P>避免数据流。应用程序通常运行为A，而考虑嵌套的条件逻辑与列：

match['Winning Team'] = np.where(match['home_team_goal'] > match['away_team_goal'],
                                 match['Home Team'],
                                 np.where(match['home_team_goal'] < match['away_team_goal'],
                                          match['Away Team'],
                                          np.nan
                                         )
                                )

<>避免数据流应用.Apple通常作为A运行，而考虑嵌套的条件逻辑与列：

match['Winning Team'] = np.where(match['home_team_goal'] > match['away_team_goal'],
                                 match['Home Team'],
                                 np.where(match['home_team_goal'] < match['away_team_goal'],
                                          match['Away Team'],
                                          np.nan
                                         )
                                )

使用，而不是应用，这应该不是一个好主意在这里的引擎盖下：

match = pd.DataFrame({
        'Home Team':list('abcdef'),
        'Away Team':list('ghijkl'),
         'home_team_goal':[14,5,4,5,5,4],
         'away_team_goal':[7,8,2,4,8,4],
})

m1 = match.home_team_goal>match.away_team_goal
m2 = match.home_team_goal<match.away_team_goal
match['winner'] = np.select([m1, m2], [match['Home Team'],  match['Away Team']], default=None)
print (match)
  Home Team Away Team  home_team_goal  away_team_goal winner
0         a         g              14               7      a
1         b         h               5               8      h
2         c         i               4               2      c
3         d         j               5               4      d
4         e         k               5               8      k
5         f         l               4               4   None

使用，而不是应用，这应该不是一个好主意在这里的引擎盖下：

match = pd.DataFrame({
        'Home Team':list('abcdef'),
        'Away Team':list('ghijkl'),
         'home_team_goal':[14,5,4,5,5,4],
         'away_team_goal':[7,8,2,4,8,4],
})

m1 = match.home_team_goal>match.away_team_goal
m2 = match.home_team_goal<match.away_team_goal
match['winner'] = np.select([m1, m2], [match['Home Team'],  match['Away Team']], default=None)
print (match)
  Home Team Away Team  home_team_goal  away_team_goal winner
0         a         g              14               7      a
1         b         h               5               8      h
2         c         i               4               2      c
3         d         j               5               4      d
4         e         k               5               8      k
5         f         l               4               4   None

还有一种方法：

与团队一起创建dataframe：

results_di = {
    "Home Team": [
        "KV Mechelen",
        "KSV Cercle Brugge",
        "RSC Anderlecht",
        "KV Mechelen",
        "SV Zulte-Waregem",
        "FC Zürich",
        "FC St. Gallen",
        "FC Vaduz",
        "Grasshopper Club Zürich",
        "BSC Young Boys",
    ],
    "Away Team": [
        "KRC Genk",
        "Club Brugge KV",
        "SV Zulte-Waregem",
        "RSC Anderlecht",
        "KSV Roeselare",
        "FC Thun",
        "FC Thun",
        "FC Luzern",
        "FC Sion",
        "FC Basel",
    ],
    "home_team_goal": [2, 1, 2, 2, 0, 3, 1, 1, 2, 4],
    "away_team_goal": [1, 3, 0, 1, 0, 3, 0, 2, 0, 3],
}

df = pd.DataFrame(results_di)
df['winner'] = 'none'
home_win = df['home_team_goal'] - df["away_team_goal"] > 0
away_win = df["away_team_goal"] - df['home_team_goal'] > 0

df.loc[home_win, "winner"] = df.loc[home_win, 'Home Team']
df.loc[away_win, "winner"] = df.loc[away_win, 'Away Team']

printdf

                 Home Team         Away Team  home_team_goal  away_team_goal  \
0              KV Mechelen          KRC Genk               2               1   
1        KSV Cercle Brugge    Club Brugge KV               1               3   
2           RSC Anderlecht  SV Zulte-Waregem               2               0   
3              KV Mechelen    RSC Anderlecht               2               1   
4         SV Zulte-Waregem     KSV Roeselare               0               0   
5                FC Zürich           FC Thun               3               3   
6            FC St. Gallen           FC Thun               1               0   
7                 FC Vaduz         FC Luzern               1               2   
8  Grasshopper Club Zürich           FC Sion               2               0   
9           BSC Young Boys          FC Basel               4               3   

                    winner  
0              KV Mechelen  
1           Club Brugge KV  
2           RSC Anderlecht  
3              KV Mechelen  
4                     none  
5                     none  
6            FC St. Gallen  
7                FC Luzern  
8  Grasshopper Club Zürich  
9           BSC Young Boys  

还有一种方法：

与团队一起创建dataframe：

results_di = {
    "Home Team": [
        "KV Mechelen",
        "KSV Cercle Brugge",
        "RSC Anderlecht",
        "KV Mechelen",
        "SV Zulte-Waregem",
        "FC Zürich",
        "FC St. Gallen",
        "FC Vaduz",
        "Grasshopper Club Zürich",
        "BSC Young Boys",
    ],
    "Away Team": [
        "KRC Genk",
        "Club Brugge KV",
        "SV Zulte-Waregem",
        "RSC Anderlecht",
        "KSV Roeselare",
        "FC Thun",
        "FC Thun",
        "FC Luzern",
        "FC Sion",
        "FC Basel",
    ],
    "home_team_goal": [2, 1, 2, 2, 0, 3, 1, 1, 2, 4],
    "away_team_goal": [1, 3, 0, 1, 0, 3, 0, 2, 0, 3],
}

df = pd.DataFrame(results_di)
df['winner'] = 'none'
home_win = df['home_team_goal'] - df["away_team_goal"] > 0
away_win = df["away_team_goal"] - df['home_team_goal'] > 0

df.loc[home_win, "winner"] = df.loc[home_win, 'Home Team']
df.loc[away_win, "winner"] = df.loc[away_win, 'Away Team']

printdf

                 Home Team         Away Team  home_team_goal  away_team_goal  \
0              KV Mechelen          KRC Genk               2               1   
1        KSV Cercle Brugge    Club Brugge KV               1               3   
2           RSC Anderlecht  SV Zulte-Waregem               2               0   
3              KV Mechelen    RSC Anderlecht               2               1   
4         SV Zulte-Waregem     KSV Roeselare               0               0   
5                FC Zürich           FC Thun               3               3   
6            FC St. Gallen           FC Thun               1               0   
7                 FC Vaduz         FC Luzern               1               2   
8  Grasshopper Club Zürich           FC Sion               2               0   
9           BSC Young Boys          FC Basel               4               3   

                    winner  
0              KV Mechelen  
1           Club Brugge KV  
2           RSC Anderlecht  
3              KV Mechelen  
4                     none  
5                     none  
6            FC St. Gallen  
7                FC Luzern  
8  Grasshopper Club Zürich  
9           BSC Young Boys  

---

@耶斯雷尔。。。条条大路通罗马，对吗？我习惯于基于集合的嵌套逻辑，类似于其他语言，如R的ifelse，SQL的case。选择是有趣的！我仍然记得我第一次在联合国大学遇到这个函数，这真是一个很好的知识；我还使用了双、三np.where，但可读性稍差@耶斯雷尔。。。条条大路通罗马，对吗？我习惯于基于集合的嵌套逻辑，类似于其他语言，如R的ifelse，SQL的case。选择是有趣的！我仍然记得我第一次在联合国大学遇到这个函数，这真是一个很好的知识；我还使用了双、三np.where，但可读性稍差；检查你的答案，我认为你在第二个过滤器中的意思是匹配[客场球队]。检查你的答案，我认为你在第二个过滤器中的意思是匹配[客场球队]。