MODULE REFERENCE

darts_naive_seasonal

anomx.models.forecasting.darts_naive_seasonal

Darts-backed forecasting model (requires the darts extra).

DartsNaiveSeasonalModel

DartsNaiveSeasonalModel(config: dict[str, Any] | None = None)

Bases: NormalityModel, Predictive

One-step-ahead naive seasonal forecasts backed by darts.

Darts is the reference ecosystem for the time-series side of anomx; this component demonstrates the integration pattern with the smallest possible darts model. Residuals against the forecast land in model_score.

Signature parameters
ParameterTypeDefault
configdict[str, Any] | NoneNone

Members

component_keyattribute
component_key = 'darts_naive_seasonal'
component_iconattribute
signatureattribute
signature = ModelSignature(structures={DataStructure.TEMPORAL_SEQUENCE}, modalities={Modality.NUMERIC}, requirements={'regular_or_known_timestamps'})
component_nameattribute
component_name = 'Darts Naive Seasonal'
component_default_configattribute
component_default_config = {'feature_columns': [], 'seasonal_period': 1}
component_config_schemaattribute
component_config_schema = {'feature_columns': {'type': 'array'}, 'seasonal_period': {'type': 'integer'}}
capabilitiesattribute
capabilities: ClassVar[ComponentCapabilities] = ComponentCapabilities(supports_unordered_samples=False, supports_sequences=True, supports_irregular_sampling=False, requires_regular_sampling=True)
preferred_viewattribute
preferred_view: ClassVar[str] = 'timeseries'
fitmethod
fit(data: Any) -> None
Signature parameters
ParameterTypeDefault
dataAnyrequired

Returns None

predictmethod
predict(data: Any) -> pd.DataFrame
Signature parameters
ParameterTypeDefault
dataAnyrequired

Returns pd.DataFrame

savemethod
save(path: str) -> None
Signature parameters
ParameterTypeDefault
pathstrrequired

Returns None

loadmethod
load(path: str) -> DartsNaiveSeasonalModel
Signature parameters
ParameterTypeDefault
pathstrrequired

Returns DartsNaiveSeasonalModel