如何复制一行但在 Pandas 中交换值

Question

我有一个数据框，我想在其中复制行但交换值。原始显示为 df1，所需为 df2:

df1

GameID     Team1ID     Team2ID     Team1Abb      Team2Abb    Team1PTS    Team2PTS
   0        1001        1023         TOR           ATL          99          86
   1        1004        1003         DAL           POR          102         100

df2

GameID     Team1ID     Team2ID     Team1Abb      Team2Abb    Team1PTS    Team2PTS
   0        1001        1023         TOR           ATL          99          86
   0        1023        1001         ATL           TOR          86          99
   1        1004        1003         DAL           POR          102         100
   1        1003        1004         POR           DAL          100         102

Answer 1

rename 和 append

我们可以通过创建映射字典来重命名数据框的列，将 Team1 列名映射到 Team2 列名，反之亦然，然后附加重命名的数据框以复制行

c1 = df.filter(like='Team1').columns
c2 = c1.str.replace('Team1', 'Team2')

swap = df.rename(columns={**dict(zip(c1, c2)), **dict(zip(c2, c1))})
df.append(swap).sort_index(ignore_index=True)

---

   GameID  Team1ID  Team2ID Team1Abb Team2Abb  Team1PTS  Team2PTS
0       0     1001     1023      TOR      ATL        99        86
1       0     1023     1001      ATL      TOR        86        99
2       1     1004     1003      DAL      POR       102       100
3       1     1003     1004      POR      DAL       100       102

Answer 2

使用 re.sub 重命名列的另一个选项：

import re
# replace Team1 with Team2 and Team2 with Team1
swappat = lambda m: '2' if m[0] == '1' else '1'
swapteam = lambda col: re.sub('(?<=^Team)[12]', swappat, col)

# append original data frame with new data frame that has name swapped
df.append(df.rename(columns=swapteam)).sort_index()

   GameID  Team1ID  Team2ID Team1Abb Team2Abb  Team1PTS  Team2PTS
0       0     1001     1023      TOR      ATL        99        86
0       0     1023     1001      ATL      TOR        86        99
1       1     1004     1003      DAL      POR       102       100
1       1     1003     1004      POR      DAL       100       102

Answer 3

#The problem can be solved by putting the individual rows of dataframe in list 
#Then the list should be updated by swapping between two adjacent values 
#then add it to the original dataframe
def swapAdjList(df):
    # as per the current problem the value of GameId should be replicated so ignoring it by minus 1
    number_of_cols=len(df.columns) -1 
    number_of_rows=len(df.index)
    number_of_iter=int(number_of_cols/2)
    result_list=[]
    for row in range(number_of_rows):
        result_list_temp=list(df.iloc[row,1:len(df.columns)])
        k=0
        for i in range(number_of_iter):
            result_list_temp[k],result_list_temp[k+1]=result_list_temp[k+1],result_list_temp[k]
            #shifting the pointer column after two coluns
            k=k+2
        #adding the gameID as per the list 
        list_gameId=int(df.iloc[row,0:1])
        result_list_temp.insert(0,list_gameId)
        result_list.append(result_list_temp)
    
    return pd.DataFrame(result_list,columns=['GameID','Team1ID','Team2ID','Team1Abb','Team2Abb','Team1PTS','Team2PTS'])

#calling function
import pandas as pd
data=[[0,1001,1023,'TOR','ATL',99,86],[1,1004,1003,'DAL','POR',102,100]]
df=pd.DataFrame(data,columns=['GameID','Team1ID','Team2ID','Team1Abb','Team2Abb','Team1PTS','Team2PTS'])
df1=swapAdjList(df)
df=df.append(df1)
df.sort_index()

Answer 4

实现预期结果的更简单的逻辑

df2 = df1.copy()
df2.columns = ["GameID", "Team2ID", "Team1ID", "Team2Abb", "Team1Ab", "Team2PTS", "Team1PTS"]
df1.append(df2).sort_index(ignore_index=True)