FoldMask

class FoldMask(first_train_timestamp: Optional[Union[str, pandas._libs.tslibs.timestamps.Timestamp]], last_train_timestamp: Union[str, pandas._libs.tslibs.timestamps.Timestamp], target_timestamps: List[Union[str, pandas._libs.tslibs.timestamps.Timestamp]])[source]

Bases: etna.core.mixins.BaseMixin

Container to hold the description of the fold mask.

Fold masks are expected to be used for backtest strategy customization.

Init FoldMask.

Parameters
  • first_train_timestamp (Optional[Union[str, pandas._libs.tslibs.timestamps.Timestamp]]) – First train timestamp, the first timestamp in the dataset if None is passed

  • last_train_timestamp (Union[str, pandas._libs.tslibs.timestamps.Timestamp]) – Last train timestamp

  • target_timestamps (List[Union[str, pandas._libs.tslibs.timestamps.Timestamp]]) – List of target timestamps

Inherited-members

Methods

set_params(**params)

Return new object instance with modified parameters.

to_dict()

Collect all information about etna object in dict.

validate_on_dataset(ts, horizon)

Validate fold mask on the dataset with specified horizon.

validate_on_dataset(ts: etna.datasets.tsdataset.TSDataset, horizon: int)[source]

Validate fold mask on the dataset with specified horizon.

Parameters