Question

我有一个数据框 UFC_data ，其中包含与A，B和战斗机的UFC战斗的清单。

data = {'fighter_A':['Tony Ferguson', 'Al Iaquinta', 'Robbie Lawler'], 
        'fighter_B':['Donald Cerrone', 'Donald Cerrone', 'Donald Cerrone'],
        'winner': ['Tony Ferguson', 'Donald Cerrone', 'Robbie Lawler']}
UFC_data = pd.DataFrame(data)

+-------+---------------+----------------+----------------+
| Index |   fighter_A   |   fighter_B    |     winner     |
+-------+---------------+----------------+----------------+
|     0 | Tony Ferguson | Donald Cerrone | Tony Ferguson  |
|     1 | Al Iaquinta   | Donald Cerrone | Donald Cerrone |
|     2 | Robbie Lawler | Donald Cerrone | Robbie Lawler  |
+-------+---------------+----------------+----------------+

我创建了第二个数据框 fighter_list ，列出了我想完成的每架战斗机，并列出了每架战斗机的获胜次数

+-------+----------------+------+
| Index |    fighter     | wins |
+-------+----------------+------+
|     0 | Tony Ferguson  |      |
|     1 | Al Iaquinta    |      |
|     2 | Robbie Lawler  |      |
|     3 | Donald Cerrone |      |
+-------+----------------+------+

直到现在我还无法返回每架战斗机的胜利次数

for index, row in fighter_list.iterrows():
    fighter_list['wins'][index] = UFC_data['winner'].loc[UFC_data['winner'] == fighter_list['fighter'][index]].sum()

Answer 1

使用Series.map创建的Series中的Series.value_counts，最后用Series.fillna替换丢失的值，并将值转换为整数：

s = UFC_data['winner'].value_counts()
fighter_list['wins'] = fighter_list['fighter'].map(s).fillna(0).astype(int)
print (fighter_list)
          fighter  wins
0   Tony Ferguson     1
1     Al Iaquinta     0
2   Robbie Lawler     1
3  Donald Cerrone     1

Answer 2

定义以下功能：

def winCnt(row):
    return UFC_data.winner[UFC_data.winner == row.fighter].count()

然后将其应用于每一行，并将结果保存在目标列中：

fighter_list['wins'] = fighter_list.apply(winCnt, axis=1)

Python DataFrame-比较是否满足条件在第二个数据框中进行计数

2 个答案: