time segments aggregate¶

path: mlstars.custom.timeseries_preprocessing.time_segments_aggregate

description: this primitive creates an equi-spaced time series by aggregating values over fixed specified interval.

argument	type	description
parameters
`X`	`numpy.ndarray` or `pandas.DataFrame`	n-dimensional sequence of values
`time_column`	`str`	column of `X` that contains time values
hyperparameters
`interval`	`int`	integer denoting time span to compute aggregation of
`method`	`str`	string describing aggregation method or list of strings describing multiple aggregation methods. If not given, mean is used
output
`X`	`numpy.ndarray`	sequence of aggregated values, one column for each aggregation method
`index`	`numpy.ndarray`	sequence of index values (first index of each aggregated segment)

In [1]: from mlstars import load_primitive

In [2]: primitive = load_primitive('mlstars.custom.timeseries_preprocessing.time_segments_aggregate',
   ...:     arguments={"time_column": "timestamp", "interval":10, "method":'mean'})
   ...: 

In [3]: df = pd.DataFrame({
   ...:     'timestamp': list(range(50)),
   ...:     'value': [1] * 50})
   ...: 

In [4]: X, index = primitive.produce(X=df)

In [5]: pd.DataFrame({"timestamp": index, "value": X[:, 0]})
Out[5]: 
   timestamp  value
0          0    1.0
1         10    1.0
2         20    1.0
3         30    1.0
4         40    1.0

Primitives intervals to mask