dlpy.timeseries.TimeseriesTable.timeseries_accumlation

TimeseriesTable.timeseries_accumlation(acc_interval='day', timeid=None, timeseries=None, groupby=None, extra_num_columns=None, default_ts_acc='sum', default_col_acc='avg', acc_method_byvar=None, user_defined_interval=None)

Accumulate the TimeseriesTable into regular consecutive intervals

Parameters:
acc_interval : string, optional

The accumulation interval, such as ‘year’, ‘qtr’, ‘month’, ‘week’, ‘day’, ‘hour’, ‘minute’, ‘second’.

timeid : string, optional

Specifies the column name for the timeid. If None, it will take the timeid specified in timeseries_formatting.
Default: None

timeseries : string or list-of-strings, optional

Specifies the column name for the timeseries, that will be part of the input or output of the RNN. If str, then it is univariate time series. If list of strings, then it is multivariate timeseries. If None, it will take the timeseries specified in timeseries_formatting.
Default: None

groupby : string or list-of-strings, optional

The groupby variables.
Default: None

extra_num_columns : string or list-of-strings, optional

Specifies the addtional numeric columns to be included for accumulation. These columns can include static feature, and might be accumulated differently than the timeseries that will be used in RNN. if None, it means no additional numeric columns will be accumulated for later processing and modeling.
Default: None

default_ts_acc : string, optional

Default accumulation method for timeseries.
Default: sum

default_col_acc : string, optional

Default accumulation method for additional numeric columns
Default: avg

acc_method_byvar : dict, optional

It specifies specific accumulation method for individual columns, if the method is different from the default. It has following structure: {‘column1 name’: ‘accumulation method1’, ‘column2 name’: ‘accumulation method2’, …}
Default: None

user_defined_interval: string, optional

Use the user-defined interval to overwrite acc_interval See more details here: https://go.documentation.sas.com/?docsetId=casforecast&docsetTarget=casforecast_tsmodel_syntax04.htm&docsetVersion=8.4

Examples

>>> from swat import CAS
>>> from dlpy.timeseries import TimeseriesTable
>>> s=CAS("cloud.example.com", 5570)
>>> time_tbl = TimeseriesTable.from_localfile(s, "path/to/file.csv", casout=dict(name='time_tbl', replace=True))
>>> time_tbl.timeseries_formatting(timeid='datetime',
...                              timeseries='series',
...                              timeid_informat='ANYDTDTM19.',
...                              timeid_format='DATETIME19.')
>>> time_tbl.timeseries_accumlation(acc_interval='day',
...                               timeseries = 'series',
...                               groupby=['id1var', 'id2var'])