Python：按小时在数据框中选择行

Question

I have the below dataframe in csv file, I would like to select all rows corresponding to current hour.

time,values

2018-10-28 08:16:49.469508,48

2018-10-28 08:16:54.471987,48

2018-10-28 08:16:59.475236,48

2018-10-28 08:17:04.478681,48

Below is the funtion I am trying current = datetime.datetime.now()

start = datetime.datetime(current.year,current.month,current.day,current.hour,0)

end = datetime.datetime(current.year,current.month,current.day,current.hour,59)

df = pd.io.parsers.read_csv('water_data1.csv', parse_dates=[0], index_col=0)

print(df.query('start < time < end'))

I get the following error

pandas.core.computation.ops.UndefinedVariableError: name 'start' is not defined

Could someone suggest what is the right syntax to achieve this. Thanks Hemanth

Answer 1

你可以试试

df[(df['time'] > start) & (df['time'] < end])]

Answer 2

pd.DataFrame.query requires external variables to be preceded by @ :

df = pd.DataFrame({'A': list(range(10))})

start, end = 3, 6

print(df.query('@start < A < @end'))

   A
4  4
5  5

You can also use pd.Series.between :

res = df[df['A'].between(start, end, inclusive=False)]

Finally, when working with datetime values, you should prefer pd.Timestamp over regular Python types:

now = pd.Timestamp('now')
start = now.replace(second=0, microsecond=0)
end = now.replace(second=59, microsecond=0)

print((start, end))

(Timestamp('2018-11-01 17:36:00'), Timestamp('2018-11-01 17:36:59'))