Python 在dataframe中匹配名称并更新目标值

我有一个像这样的数据框

               Player   Pos Team    Opp Def v Pos  Salary  StdDev Own  My Proj  Target_Score
0          Jared Goff  CPTN  LAR  vs NE       7th   16500     9.0  0%    27.41           NaN
1          Cam Newton  CPTN   NE  @ LAR       2nd   15900     8.9  0%    26.75           NaN
2        Robert Woods  CPTN  LAR  vs NE       8th   13800     8.6  0%    22.68           NaN
3         Cooper Kupp  CPTN  LAR  vs NE       8th   14400     9.2  0%    21.84           NaN
4       Jakobi Meyers  CPTN   NE  @ LAR       2nd   10200     7.0  0%    21.59           NaN
5          Jared Goff    QB  LAR  vs NE       7th   11000     9.0  0%    18.27          33.2
6          Cam Newton    QB   NE  @ LAR       2nd   10600     8.9  0%    17.83          33.2
7           Cam Akers  CPTN  LAR  vs NE      22nd   12000     4.8  0%    17.82           NaN
8       Damien Harris  CPTN   NE  @ LAR       4th   13200     5.2  0%    16.16           NaN
9        Robert Woods    WR  LAR  vs NE       8th    9200     8.6  0%    15.12          30.7
10        Cooper Kupp    WR  LAR  vs NE       8th    9600     9.2  0%    14.56          30.7
11      Jakobi Meyers    WR   NE  @ LAR       2nd    6800     7.0  0%    14.39          28.1
12       Damiere Byrd  CPTN   NE  @ LAR       2nd    1200     6.6  0%    13.59           NaN
13           Matt Gay  CPTN  LAR  vs NE      25th    5700     3.4  0%    13.35           NaN
14        James White  CPTN   NE  @ LAR       4th    9600     7.5  0%    12.75           NaN
15     Gerald Everett  CPTN  LAR  vs NE       9th    5100     5.2  0%    12.32           NaN
16          Cam Akers    RB  LAR  vs NE      22nd    8000     4.8  0%    11.88          30.0
17       Tyler Higbee  CPTN  LAR  vs NE       9th    7500     6.7  0%    11.60           NaN
18          Nick Folk  CPTN   NE  @ LAR       5th    6000     3.7  0%    11.46           NaN
19               Rams  CPTN  LAR  vs NE      27th    6600     6.8  0%    11.24           NaN
20      Josh Reynolds  CPTN  LAR  vs NE       8th    8400     5.7  0%    11.21           NaN
21       N Keal Harry  CPTN   NE  @ LAR       2nd    2700     4.3  0%    10.85           NaN
22      Damien Harris    RB   NE  @ LAR       4th    8800     5.2  0%    10.77          30.0
23       Damiere Byrd    WR   NE  @ LAR       2nd     800     6.6  0%     9.06          10.0
24           Matt Gay     K  LAR  vs NE      25th    3800     3.4  0%     8.90           NaN
25        James White    RB   NE  @ LAR       4th    6400     7.5  0%     8.50          29.0
26     Gerald Everett    TE  LAR  vs NE       9th    3400     5.2  0%     8.21          24.0
27  Darrell Henderson  CPTN  LAR  vs NE      22nd   11100     7.1  0%     7.92           NaN
28       Tyler Higbee    TE  LAR  vs NE       9th    5000     6.7  0%     7.73          26.3
29          Nick Folk     K   NE  @ LAR       5th    4000     3.7  0%     7.64           NaN
30           Patriots  CPTN   NE  @ LAR      10th    6300     6.8  0%     7.59           NaN
31               Rams   DST  LAR  vs NE      27th    4400     6.8  0%     7.49          15.3
32      Josh Reynolds    WR  LAR  vs NE       8th    5600     5.7  0%     7.47          28.1

我需要迭代每个玩家并匹配玩家名称，然后更新Target_Score列

例如：

匹配Jared Goff CPTN和Jared Goff QB

for index, row in player_df.iterrows():
    # if player name with position CPTN == player name with position QB
    # player name with position CPTN Target_Score = player name with position 
    # QB['Target_Score'] * 1.5

---

基本上，我需要将每个玩家与其对应的CPTN位置进行匹配，并迭代更新CPTN目标分数，使其成为位置目标分数*1.5。

选择作为基本事实的分数，对其进行操作，然后将其映射回CPTN位置。这假设玩家只在CPTN和非CPTN位置出现一次

cptn_selector = df["Pos"] == "CPTN"
other_selector = ~cptn_selector
players_and_scores = df.loc[other_selector, ["Player", "Target_Score"]]
players_and_scores["Target_Score_CPTN"] = players_and_scores["Target_Score"].values * 1.5
name_to_score_mapping = pd.Series(
    players_and_scores["Target_Score"].values,
    index=players_and_scores["Player"]
).to_dict()
df.loc[cptn_selector, "Target_Score"] = df.loc[cptn_selector, "Player"].map(name_to_score_mapping)

文件化内联

# Some test data
player_df = pd.DataFrame({ 'Player': ['Jared Goff', 'Cam Newton']*2,
                    'Pos': ['CPTN','CPTN', 'QB', 'WR'],
                    'Target_Score': [0,1,2,3]}) 

new_player_df = player_df.copy()
print (new_player_df)

print ("*"*10)
# Iterate overs players with CPTN 
for index, row in player_df[player_df['Pos'] == 'CPTN'].iterrows():
  # Get the match
  match = player_df[(player_df['Player'] == row['Player']) & 
                    (player_df['Pos'] != 'CPTN')]
  # If there is a match
  if len(match) > 0:
    # Repace the score
    new_player_df.loc[index, 'Target_Score'] = match.iloc[0]['Target_Score'] * 1.5

print (new_player_df)

输出：

       Player   Pos  Target_Score
0  Jared Goff  CPTN             0
1  Cam Newton  CPTN             1
2  Jared Goff    QB             2
3  Cam Newton    WR             3
**********
       Player   Pos  Target_Score
0  Jared Goff  CPTN           3.0
1  Cam Newton  CPTN           4.5
2  Jared Goff    QB           2.0
3  Cam Newton    WR           3.0

---

我对每个职位都需要这个，而不仅仅是QBSo，我假设每个付款人除了CPTN之外，只有更多的职位，因为如果有多个职位，那么

'Target\u分数将被覆盖，并且只有最后的匹配才会出现在结果中。如果是这种情况，请参阅我的编辑。是的，每个位置QB、RB、WR、TE只有一个CPTN，DST@AustinJohnson然后请检查编辑后的答案。这样我就得到了相同的分数。这不是用这一行乘以1.5试试看
players_和_分数[“Target_分数\u CPTN”]=players_和_分数[“Target_分数”]。值*1.5