LagsAdder

class openstef_models.transforms.time_domain.lags_adder.LagsAdder(**data: Any) → None

Bases: BaseConfig, TimeSeriesTransform

Transform that adds lag features to time series data.

Creates lagged copies of the target variable to capture temporal dependencies. Handles both single-horizon and multi-horizon (versioned) datasets, ensuring lags are valid for each forecast horizon.

Invariants

• fit() must be called before transform() when add_autocorr_lags=True

• Lags are only added if they fall within the available history window

• For each horizon, only lags >= horizon are included (prevents data leakage)

• All lag features use consistent naming: {target_column}_lag_{duration}

Parameters:

data (Any)

history_available: timedelta

horizons: list[LeadTime]

custom_lags: list[timedelta] | None

target_column: str

max_day_lags: int

add_trivial_lags: bool

add_autocorr_lags: bool

lag_fallback_offset: timedelta | None

property horizon_lags: dict[LeadTime, list[timedelta]]

Mapping of forecast horizons to their valid lag features.

property lags: list[timedelta]

All lag durations configured for this transform, sorted descending.

property min_horizon: timedelta

Shortest forecast horizon to determine minimum lag requirements.

model_post_init(context: Any) → None

Override this method to perform additional initialization after __init__ and model_construct. This is useful if you want to do some validation that requires the entire model to be initialized.

Parameters:

context (Any)

Return type:

None

property is_fitted: bool

Check if the transform has been fitted.

fit(data: TimeSeriesDataset) → None

Fit the transform to the input data.

This method should be called before applying the transform to the data. It allows the transform to learn any necessary parameters from the data.

Parameters:

• data (TimeSeriesDataset) – The input data to fit the transform on.

• data

Return type:

None

transform(data: TimeSeriesDataset) → TimeSeriesDataset

Transform the input data.

This method should apply a transformation to the input data and return a new instance.

Parameters:

• data (TimeSeriesDataset) – The input data to be transformed.

• data

Returns:

A new instance of the transformed data.

Raises:

NotFittedError – If the transform has not been fitted yet.

Return type:

TimeSeriesDataset

features_added() → list[str]

List of feature names added by this transform.

Return type:

list[str]

Returns:

A list of strings representing the names of features added to the dataset by this transform. Default is an empty list.

model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': False, 'extra': 'ignore', 'protected_namespaces': ()}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].