LassoCVModel

Lasso regression with cross-validated regularization using scikit-learn’s LassoCV.

Basic Usage

from epftoolbox2.models import LassoCVModel

model = LassoCVModel(
    predictors=[
        "load_actual",
        "is_monday_d+{horizon}",
        "is_tuesday_d+{horizon}",
        "is_wednesday_d+{horizon}",
        "is_thursday_d+{horizon}",
        "is_friday_d+{horizon}",
        "is_saturday_d+{horizon}",
        "is_sunday_d+{horizon}",
        "is_holiday_d+{horizon}",
        "daylight_hours_d+{horizon}",
    ],
    training_window=365,
    cv=5,
    name="LassoCV",
)

Parameters

Parameter	Type	Default	Description
`predictors`	List	Required	Predictor specifications
`training_window`	int	`365`	Days of training data
`cv`	int	`5`	Cross-validation folds
`max_iter`	int	`10000`	Max optimization iterations
`name`	str	`"Model"`	Display name for reports

When to Use

Many predictors: Lasso performs automatic feature selection
Correlated features: Regularization handles multicollinearity
Preventing overfitting: L1 penalty shrinks coefficients

Example

from epftoolbox2.pipelines import ModelPipeline
from epftoolbox2.models import LassoCVModel
from epftoolbox2.evaluators import MAEEvaluator
from epftoolbox2.exporters import TerminalExporter

predictors = [
    "load_actual",
    "is_monday_d+{horizon}",
    "is_tuesday_d+{horizon}",
    "is_wednesday_d+{horizon}",
    "is_thursday_d+{horizon}",
    "is_friday_d+{horizon}",
    "is_saturday_d+{horizon}",
    "is_sunday_d+{horizon}",
    "is_holiday_d+{horizon}",
    "daylight_hours_d+{horizon}",
    *[f"load_actual_h-{i}" for i in range(1, 169)],
]

pipeline = (
    ModelPipeline()
    .add_model(LassoCVModel(predictors=predictors, cv=5, name="LassoCV"))
    .add_evaluator(MAEEvaluator())
    .add_exporter(TerminalExporter())
)

report = pipeline.run(data=df, test_start="2024-02-01", test_end="2024-03-01")