Function reference

Preprocessing

Data curation, outlier detection, and split tools.

preprocessing._binner	QuantileBinner class for binning data into quantile-based bins.
preprocessing._common	Common preprocessing functions and utilities.
preprocessing._differentiator	Time series differentiator transformer.
preprocessing._rolling
preprocessing.outlier	Outlier detection utilities (legacy wrapper for spotforecast2_safe).
preprocessing.outlier_plots
preprocessing.split
preprocessing.time_series_visualization	Time series visualization.

Search algorithms and cross-validation tools.

model_selection.bayesian_search	Bayesian hyperparameter search functions for forecasters using Optuna.
model_selection.grid_search
model_selection.random_search	Random search hyperparameter optimization for forecasters.
model_selection.split_base	Base class for time series cross-validation splitting.
model_selection.split_ts_cv	Time series cross-validation splitting.
model_selection.spotoptim_search	Hyperparameter search functions for forecasters using SpotOptim.
model_selection.utils_common	Common validation and initialization utilities for model selection.
model_selection.utils_metrics	Metrics calculation utilities for model selection.

Model management and configurators.

manager.models.forecaster_recursive_model_full	Full-featured base forecasting model with Bayesian tuning and SHAP.
manager.models.forecaster_recursive_lgbm_full	LGBM forecaster with real Bayesian tuning and SHAP.
manager.models.forecaster_recursive_xgb_full	XGBoost forecaster with real Bayesian tuning and SHAP.
manager.plotter	Module for generating interactive prediction plots.
manager.trainer_full	Module for managing full model training.
manager.multitask.agg_predictor	Aggregate per-target prediction packages into a weighted forecast.
manager.multitask.BaseTask	Shared base for all multi-target forecasting pipeline tasks.
manager.multitask.LazyTask	Task 1 — Lazy Fitting with default LightGBM parameters.
manager.multitask.OptunaTask	Task 3 — Optuna Bayesian hyperparameter tuning.
manager.multitask.SpotOptimTask	Task 4 — SpotOptim surrogate-model Bayesian tuning.
manager.multitask.PredictTask	Task 5 — Predict-only using previously saved models.
manager.multitask.CleanTask	Cache-cleaning task — removes all cached data from the pipeline cache.
manager.multitask.MultiTask	Orchestrates a multi-target time-series forecasting pipeline.
manager.multitask.run	Run the MultiTask forecasting pipeline and return predictions.

Forecasting utilities and metrics.

forecaster.metrics	Metrics for evaluating forecasting models.
forecaster.utils
forecaster.recursive._warnings

Statistical analysis tools.

Demonstration and predefined forecasting tasks.

tasks.task_demo	Task demo: compare baseline, covariate, and custom LightGBM forecasts against ground truth.
tasks.task_entsoe	Unified CLI task script for ENTSO-E data downloading, model training, and prediction.
tasks.task_n_to_1
tasks.task_n_to_1_dataframe
tasks.task_n_to_1_with_covariates	N-to-1 Forecasting with Exogenous Covariates and Prediction Aggregation.
tasks.task_n_to_1_with_covariates_and_dataframe	N-to-1 Forecasting with Exogenous Covariates and Prediction Aggregation.

Validation and data transformation utilities.

utils.data_transform	Data transformation utilities for time series forecasting.
utils.forecaster_config	Forecaster configuration utilities.
utils.generate_holiday	Utilities for generating holiday dataframe as covariate.
utils.validation	Validation utilities for time series forecasting.

Spotforecast exception classes.

Custom exceptions and warnings for spotforecast2.