Data

Utilies for generating time series datasets

---

source

generate_series

 generate_series (n_series:int, freq:str='D', min_length:int=50,
                  max_length:int=500, n_static_features:int=0,
                  equal_ends:bool=False, with_trend:bool=False,
                  static_as_categorical:bool=True, n_models:int=0,
                  level:Optional[List[float]]=None,
                  engine:Literal['pandas','polars']='pandas', seed:int=0)

Generate Synthetic Panel Series.

	Type	Default	Details
n_series	int		Number of series for synthetic panel.
freq	str	D	Frequency of the data (pandas alias). Seasonalities are implemented for hourly, daily and monthly.
min_length	int	50	Minimum length of synthetic panel’s series.
max_length	int	500	Maximum length of synthetic panel’s series.
n_static_features	int	0	Number of static exogenous variables for synthetic panel’s series.
equal_ends	bool	False	Series should end in the same timestamp.
with_trend	bool	False	Series should have a (positive) trend.
static_as_categorical	bool	True	Static features should have a categorical data type.
n_models	int	0	Number of models predictions to simulate.
level	Optional	None	Confidence level for intervals to simulate for each model.
engine	Literal	pandas	Output Dataframe type.
seed	int	0	Random seed used for generating the data.
Returns	Union		Synthetic panel with columns [unique_id, ds, y] and exogenous features.

synthetic_panel = generate_series(n_series=2)
synthetic_panel.groupby('unique_id', observed=True).head(4)

	unique_id	ds	y
0	0	2000-01-01	0.357595
1	0	2000-01-02	1.301382
2	0	2000-01-03	2.272442
3	0	2000-01-04	3.211827
222	1	2000-01-01	5.399023
223	1	2000-01-02	6.092818
224	1	2000-01-03	0.476396
225	1	2000-01-04	1.343744