datetime - pandas dataframe getting daily/weekly/hourly data -
i have pandas dataframe index = datetime.datetime(year,month,day,hour,minute)
i want able use hourly/ daily /weekly data hourly data last entry corresponding every hour in frame.
is there inbuilt way this? tried making cases e.g. in case of daily data, changed hour , minute entry zero, still have dataframe multiple entries same day. how can last entry corresponding each day?
sample dataframe:
index x y 2016-01-01 00:07:00-05:00 1.000 0.000 2016-01-01 00:10:00-05:00 1.000 0.000 2016-01-01 00:15:00-05:00 1.000 0.000 2016-01-01 00:16:00-05:00 1.000 0.000 2016-01-01 00:20:00-05:00 1.000 0.000 2016-01-01 00:21:00-05:00 1.000 0.000 2016-01-01 00:26:00-05:00 1.000 0.000 2016-01-01 00:31:00-05:00 1.000 0.000 2016-01-01 00:37:00-05:00 1.000 0.000 2016-01-01 00:40:00-05:00 1.000 0.000 2016-01-01 00:46:00-05:00 1.000 0.000 2016-01-01 00:51:00-05:00 1.000 0.000 2016-01-01 00:56:00-05:00 1.000 0.000 2016-01-03 19:26:00-05:00 1.000 0.000 2016-01-03 19:34:00-05:00 1.000 0.000 2016-01-03 20:02:00-05:00 1.000 0.000 2016-01-03 20:06:00-05:00 1.000 0.000 2016-01-03 20:07:00-05:00 1.000 0.000 2016-01-03 20:08:00-05:00 1.000 0.000 2016-01-03 20:10:00-05:00 1.000 0.000 2016-01-03 20:11:00-05:00 1.000 0.000 2016-01-03 20:12:00-05:00 1.000 0.000 2016-01-03 20:13:00-05:00 1.000 0.000
assuming understand question (it helpful see code examples), sounds use resample:
df.resample('d', how='sum')
it works groupby or pivot table:
dataframe.resample(rule, how=none, axis=0, fill_method=none, closed=none, label=none, convention='start', kind=none, loffset=none, limit=none, base=0) convenience method frequency conversion , resampling of regular time-series data.
parameters: rule : string offset string or object representing target conversion axis : int, optional, default 0 closed : {‘right’, ‘left’} side of bin interval closed label : {‘right’, ‘left’} bin edge label label bucket convention : {‘start’, ‘end’, ‘s’, ‘e’} loffset : timedelta adjust resampled time labels base : int, default 0 frequencies evenly subdivide 1 day, “origin” of aggregated intervals. example, ‘5min’ frequency, base range 0 through 4. defaults 0
Comments
Post a Comment