Python：如何读取列表中的所有元素并从dataframe检索相应的值

我有一个数据帧

df

，如下所示：

loc     end_time            ts          file
TPHD    2019-06-03 16:45:30 43619.4375  trial.csv
TPCL    2019-06-03 16:30:00 43619.5520  trial.csv
TPHD    2019-06-03 16:15:30 43619.6774  trial.csv
TPBL    2019-06-03 16:15:30 43619.4479  trial.csv
TPBL    2019-06-03 14:43:45 43619.6982  mgrflash.csv
TPCL    2019-06-03 13:15:00 43619.4375  mgrflash.csv
TPCL    2019-06-03 11:15:30 43619.6875  mgrflash.csv
TPCL    2019-06-03 10:45:00 43619.6137  trial.csv
TPBL    2019-06-03 10:30:00 43619.6774  mgrflash.csv
TPHD    2019-06-03 10:30:00 43619.4690  mgrflash.csv

df_test = df.drop_duplicates(['loc','file'])
location = ['TPCL','TPBL','TPHD']
now_dt = dt.now()
for i in location:
    if i in df_test['loc'].tolist():
        t_update_loc = df_test.loc[df_test['loc']==i,'end_time']
        d = []
        for j in t_update_loc.tolist():
            diff = now_dt - j
            d.append(diff)

loc    time_stamp
TPCL   2019-06-03 16:30:00
TPBL   2019-06-03 16:15:30
TPHD   2019-06-03 16:45:30

目标：我想找到每个位置和文件的

end\u time

和指定时间（如

datetime.datetime.now（）

）之间的时差。为此，我尝试了以下方法：

loc     end_time            ts          file
TPHD    2019-06-03 16:45:30 43619.4375  trial.csv
TPCL    2019-06-03 16:30:00 43619.5520  trial.csv
TPHD    2019-06-03 16:15:30 43619.6774  trial.csv
TPBL    2019-06-03 16:15:30 43619.4479  trial.csv
TPBL    2019-06-03 14:43:45 43619.6982  mgrflash.csv
TPCL    2019-06-03 13:15:00 43619.4375  mgrflash.csv
TPCL    2019-06-03 11:15:30 43619.6875  mgrflash.csv
TPCL    2019-06-03 10:45:00 43619.6137  trial.csv
TPBL    2019-06-03 10:30:00 43619.6774  mgrflash.csv
TPHD    2019-06-03 10:30:00 43619.4690  mgrflash.csv

df_test = df.drop_duplicates(['loc','file'])
location = ['TPCL','TPBL','TPHD']
now_dt = dt.now()
for i in location:
    if i in df_test['loc'].tolist():
        t_update_loc = df_test.loc[df_test['loc']==i,'end_time']
        d = []
        for j in t_update_loc.tolist():
            diff = now_dt - j
            d.append(diff)

loc    time_stamp
TPCL   2019-06-03 16:30:00
TPBL   2019-06-03 16:15:30
TPHD   2019-06-03 16:45:30

上述代码显示的结果不正确。事实上，它只拾取

TPHD

值。以上的输出如下：

_libs.tslibs.timedeltas.Timedelta  1     140 days 02:55:06.056170
_libs.tslibs.timedeltas.Timedelta  1     140 days 09:10:36.056170

理想情况下，我希望数据帧如下所示：

loc     end_time            ts          file
TPHD    2019-06-03 16:45:30 43619.4375  trial.csv
TPCL    2019-06-03 16:30:00 43619.5520  trial.csv
TPHD    2019-06-03 16:15:30 43619.6774  trial.csv
TPBL    2019-06-03 16:15:30 43619.4479  trial.csv
TPBL    2019-06-03 14:43:45 43619.6982  mgrflash.csv
TPCL    2019-06-03 13:15:00 43619.4375  mgrflash.csv
TPCL    2019-06-03 11:15:30 43619.6875  mgrflash.csv
TPCL    2019-06-03 10:45:00 43619.6137  trial.csv
TPBL    2019-06-03 10:30:00 43619.6774  mgrflash.csv
TPHD    2019-06-03 10:30:00 43619.4690  mgrflash.csv

df_test = df.drop_duplicates(['loc','file'])
location = ['TPCL','TPBL','TPHD']
now_dt = dt.now()
for i in location:
    if i in df_test['loc'].tolist():
        t_update_loc = df_test.loc[df_test['loc']==i,'end_time']
        d = []
        for j in t_update_loc.tolist():
            diff = now_dt - j
            d.append(diff)

loc    time_stamp
TPCL   2019-06-03 16:30:00
TPBL   2019-06-03 16:15:30
TPHD   2019-06-03 16:45:30

---

如何获取上述数据帧

如果我理解得很好，您可以简单地使用：

---

df['diff']=datetime.datetime.now（）-df.end\u time

假设

end\u time

是一个datetime对象

导入日期时间

---

df=df.drop\u duplicates（['loc']）。assign（time\u stamp=lambda x:datetime.datetime.now（）-x['end\u time']）

没问题。但是如何获得上面的数据帧。尤其是在运行位置中的for i循环时，为什么没有正确读取

位置

列表为什么需要for循环？在您的预期输出中，为什么每个位置只有一行？我以为你想要一个地址和文件名？请进一步解释如何获得输出df