Forecaster

Prophetverse

API-Referenz Quellcode

Categorical in XInsamplePred intPred int insampleExogenous

Univariate prophetverse forecaster - prophet model implemented in numpyro.

Estimator from the prophetverse package by felipeangelimvieira.

Differences to facebook’s prophet:

logistic trend. Here, another parametrization is considered, and the capacity is not passed as input, but inferred from the data.
the users can pass arbitrary sktime transformers as feature_transformer, for instance FourierFeatures or HolidayFeatures.
no default weekly_seasonality/yearly_seasonality, this is left to the user via the feature_transformer parameter
Uses changepoint_interval instead of n_changepoints to set

changepoints.

accepts configurations where each exogenous variable has a different function relating it to its additive effect on the time series. One can, for example, set different priors for a group of feature, or use a Hill function to model the effect of a feature.

Schnellstart

python

from sktime.forecasting.prophetverse import Prophetverse

estimator = Prophetverse(trend='linear', exogenous_effects=None, default_effect=None, feature_transformer=None, noise_scale=None, likelihood='normal', scale=None, rng_key=None, inference_engine=None, broadcast_mode='estimator')

Parameter(9)

trendUnion[str, BaseEffect], optional

Type of trend to use. Either “linear” (default) or “logistic”, or a custom effect object.

exogenous_effectsOptional[List[BaseEffect]], optional

List of effect objects defining the exogenous effects.

default_effectOptional[BaseEffect], optional

The default effect for variables without a specified effect.

feature_transformersktime transformer, optional

Transformer object to generate additional features (e.g.,: Fourier terms).

noise_scalefloat, optional

Scale parameter for the observation noise. Must be greater than 0. (default: 0.05)

likelihoodstr, optional

The likelihood model to use. One of “normal”, “gamma”, or: “negbinomial”. (default: “normal”)

scaleoptional

Scaling value inferred from the data.

rng_keyoptional

A jax.random.PRNGKey instance, or None.

inference_engineoptional

An inference engine for running the model.

Beispiele

>>> from sktime.datasets import load_airline
>>> from sktime.forecasting.prophetverse import Prophetverse
>>> from prophetverse.effects.fourier import LinearFourierSeasonality
>>> from prophetverse.utils.regex import no_input_columns
>>> y = load_airline ()
>>> model = Prophetverse (
... exogenous_effects = [
... (
... "seasonality",
... LinearFourierSeasonality (
... sp_list = [12 ],
... fourier_terms_list = [3 ],
... freq = "M",
... effect_mode = "multiplicative",
... ),
... no_input_columns,
... )
... ],
... )
>>> model. fit (y)
>>> model. predict (fh = [1, 2, 3 ])