ValueError:您尝试在datetime64 [ns]和对象列上合并。如果要继续,则应使用pd.concat

时间:2018-08-31 04:59:28

标签: python pandas

我想制作一个数据透视表并将其填充到Date中。我编写了代码。

user_dct ={100:"Tom",101:"Jon",102:"Daisy"}
for key,value in user_dct.items():
         file= './'+value+'.csv'
       df = pd.read_csv(file)

每个df是

    Date        ID        Name    Score        Rank
0    2011-01-12  100     Tom        40            C
1    2011-01-14  100     Tom        60            B
2    2011-01-19  100     Tom        80            A
・
・
・


   Date        ID        Name    Score        Rank
0    2011-01-12  101     Jon        30            C
1    2011-01-14  101     Jon        50            C
2    2011-01-19  101     Jon        60            B
・
・
・

user_dct ={100:"Tom",101:"Jon",102:"Daisy"}
dfs = []
for key,value in user_dct.items():
    file= './'+value+'.csv'
    dfs.append(pd.read_csv(file, parse_dates=['Date']))

df = pd.concat(dfs, ignore_index=True)
df =df.sort_values(['Date','ID']).set_index(['Date','ID'])

date_df = pd.DataFrame({'Date':pd.date_range('2011-01-01','2011-12-31',freq='1D').strftime('%Y-%m-%d')})
df = pd.merge(df, date_df, on='Date', how='outer').fillna(0)

我的理想输出是

                        Name    Score        Rank
    Date      ID  
2011-01-01  100         Tom       0              0
                    101         Jon        0              0
                    102     Daisy       0              0  
                    ・
                                         ・
                                         ・
2011-01-12  100         Tom       40            C
                    101         Jon        30            C
                    102     Daisy       90            S
2011-01-14  100     Tom         60            B
                    101     Jon           50            C
                     102     Daisy       90            S
2011-01-19  100     Tom         80            A
                     101     Jon          60            B
                     102     Daisy      80            A
・
・
・

我的代码有什么问题?应如何解决?为什么会发生int类型错误?我更改了sort_values&set_index,但错误并未消失。

1 个答案:

答案 0 :(得分:2)

我认为需要:

SELECT E.ID, E.EnrolmentDate, A.NoOfMonths
FROM dbo.Enrolments E
OUTER APPLY
(
  SELECT DATEDIFF(mm, E.EnrolmentDate, MIN(A.AppointmentDate)) AS NoOfMonths
  FROM dbo.Appointments A
  WHERE A.EnrolmentId = E.ID
) A