multitask.factories.quantile_lgbm_forecaster_factory

multitask.factories.quantile_lgbm_forecaster_factory(
    config,
    *,
    quantiles=DEFAULT_QUANTILES,
    weight_func=None,
    target=None,
)

Return one quantile-regression LightGBM ForecasterRecursive per quantile.

Each forecaster uses LGBMRegressor(objective="quantile", alpha=q) and the same lag/rolling configuration as :func:default_lgbm_forecaster_factory, so a caller can fit the lower/median/upper heads independently and assemble a band with :func:predict_quantile_band. Deterministic given config.random_state; sf2-safe (LightGBM only, no torch/optuna). Refs hong16b, roma19a.

Parameters

Name	Type	Description	Default
config	Any	Object satisfying the `PipelineConfig` protocol; reads `random_state`, `lags_consider`, `window_size`.	required
quantiles	Sequence[float]	Quantile levels in the open interval `(0, 1)`. Defaults to `(0.1, 0.5, 0.9)`.	`DEFAULT_QUANTILES`
weight_func	Optional[Any]	Optional per-sample weight function.	`None`
target	Optional[str]	Accepted and ignored (parity with the default factory).	`None`

Returns

Name	Type	Description
	Dict[float, `ForecasterRecursive`]	Dict[float, ForecasterRecursive]: Map from quantile level to a fresh,
	Dict[float, `ForecasterRecursive`]	unfitted forecaster, in ascending quantile order.

Raises

Name	Type	Description
	ValueError	If quantiles is empty, out of `(0, 1)`, or has duplicates.

Examples

import types
from spotforecast2_safe.multitask.factories import (
    quantile_lgbm_forecaster_factory,
)

config = types.SimpleNamespace(
    random_state=42, lags_consider=[1, 2, 3], window_size=3
)
heads = quantile_lgbm_forecaster_factory(config, quantiles=[0.1, 0.5, 0.9])
print(sorted(heads))
print(heads[0.1].regressor.get_params()["objective"])
print(heads[0.1].regressor.get_params()["alpha"])

[0.1, 0.5, 0.9]
quantile
0.1

/home/runner/work/spotforecast2-safe/spotforecast2-safe/src/spotforecast2_safe/forecaster/base.py:476: FutureWarning: The `regressor` attribute is deprecated and will be removed in future versions. Use `estimator` instead.
  warnings.warn(