utils

Classes

PytorchForecastingDatasetBuilder([...])	Builder for PytorchForecasting dataset.
PytorchForecastingMixin()	Mixin for Pytorch Forecasting models.
_DeepCopyMixin()	Mixin for __deepcopy__ behaviour overriding.

class PytorchForecastingDatasetBuilder(max_encoder_length: int = 30, min_encoder_length: Optional[int] = None, min_prediction_idx: Optional[int] = None, min_prediction_length: Optional[int] = None, max_prediction_length: int = 1, static_categoricals: Optional[List[str]] = None, static_reals: Optional[List[str]] = None, time_varying_known_categoricals: Optional[List[str]] = None, time_varying_known_reals: Optional[List[str]] = None, time_varying_unknown_categoricals: Optional[List[str]] = None, time_varying_unknown_reals: Optional[List[str]] = None, variable_groups: Optional[Dict[str, List[int]]] = None, constant_fill_strategy: Optional[Dict[str, Union[str, float, int, bool]]] = None, allow_missing_timesteps: bool = True, lags: Optional[Dict[str, List[int]]] = None, add_relative_time_idx: bool = True, add_target_scales: bool = True, add_encoder_length: Union[bool, str] = True, target_normalizer: Union[pytorch_forecasting.data.encoders.TorchNormalizer, pytorch_forecasting.data.encoders.NaNLabelEncoder, pytorch_forecasting.data.encoders.EncoderNormalizer, str, List[Union[pytorch_forecasting.data.encoders.TorchNormalizer, pytorch_forecasting.data.encoders.NaNLabelEncoder, pytorch_forecasting.data.encoders.EncoderNormalizer]], Tuple[Union[pytorch_forecasting.data.encoders.TorchNormalizer, pytorch_forecasting.data.encoders.NaNLabelEncoder, pytorch_forecasting.data.encoders.EncoderNormalizer]]] = 'auto', categorical_encoders: Optional[Dict[str, pytorch_forecasting.data.encoders.NaNLabelEncoder]] = None, scalers: Optional[Dict[str, Union[sklearn.preprocessing._data.StandardScaler, sklearn.preprocessing._data.RobustScaler, pytorch_forecasting.data.encoders.TorchNormalizer, pytorch_forecasting.data.encoders.EncoderNormalizer]]] = None)

Builder for PytorchForecasting dataset.

Init dataset builder.

Parameters here is used for initialization of pytorch_forecasting.data.timeseries.TimeSeriesDataSet object.

Parameters

• max_encoder_length (int) –

• min_encoder_length (Optional[int]) –

• min_prediction_idx (Optional[int]) –

• min_prediction_length (Optional[int]) –

• max_prediction_length (int) –

• static_categoricals (Optional[List[str]]) –

• static_reals (Optional[List[str]]) –

• time_varying_known_categoricals (Optional[List[str]]) –

• time_varying_known_reals (Optional[List[str]]) –

• time_varying_unknown_categoricals (Optional[List[str]]) –

• time_varying_unknown_reals (Optional[List[str]]) –

• variable_groups (Optional[Dict[str, List[int]]]) –

• constant_fill_strategy (Optional[Dict[str, Union[str, float, int, bool]]]) –

• allow_missing_timesteps (bool) –

• lags (Optional[Dict[str, List[int]]]) –

• add_relative_time_idx (bool) –

• add_target_scales (bool) –

• add_encoder_length (Union[bool, str]) –

• target_normalizer (Union[pytorch_forecasting.data.encoders.TorchNormalizer, pytorch_forecasting.data.encoders.NaNLabelEncoder, pytorch_forecasting.data.encoders.EncoderNormalizer, str, List[Union[pytorch_forecasting.data.encoders.TorchNormalizer, pytorch_forecasting.data.encoders.NaNLabelEncoder, pytorch_forecasting.data.encoders.EncoderNormalizer]], Tuple[Union[pytorch_forecasting.data.encoders.TorchNormalizer, pytorch_forecasting.data.encoders.NaNLabelEncoder, pytorch_forecasting.data.encoders.EncoderNormalizer]]]) –

• categorical_encoders (Optional[Dict[str, pytorch_forecasting.data.encoders.NaNLabelEncoder]]) –

• scalers (Optional[Dict[str, Union[sklearn.preprocessing._data.StandardScaler, sklearn.preprocessing._data.RobustScaler, pytorch_forecasting.data.encoders.TorchNormalizer, pytorch_forecasting.data.encoders.EncoderNormalizer]]]) –

create_inference_dataset(ts: etna.datasets.tsdataset.TSDataset, horizon: int) → pytorch_forecasting.data.timeseries.TimeSeriesDataSet

Create inference dataset.

This method should be used only after create_train_dataset that is used during model training.

Parameters

• ts (etna.datasets.tsdataset.TSDataset) – Time series dataset.

• horizon (int) – Size of prediction to make.

Raises

ValueError: – if method was used before create_train_dataset

Return type

pytorch_forecasting.data.timeseries.TimeSeriesDataSet

create_train_dataset(ts: etna.datasets.tsdataset.TSDataset) → pytorch_forecasting.data.timeseries.TimeSeriesDataSet

Create train dataset.

Parameters

ts (etna.datasets.tsdataset.TSDataset) – Time series dataset.

Return type

pytorch_forecasting.data.timeseries.TimeSeriesDataSet

set_params(**params: dict) → etna.core.mixins.TMixin

Return new object instance with modified parameters.

Method also allows to change parameters of nested objects within the current object. For example, it is possible to change parameters of a model in a Pipeline.

Nested parameters are expected to be in a <component_1>.<...>.<parameter> form, where components are separated by a dot.

Parameters

• **params – Estimator parameters

• self (etna.core.mixins.TMixin) –

• params (dict) –

Returns

New instance with changed parameters

Return type

etna.core.mixins.TMixin

Examples

>>> from etna.pipeline import Pipeline
>>> from etna.models import NaiveModel
>>> from etna.transforms import AddConstTransform
>>> model = model=NaiveModel(lag=1)
>>> transforms = [AddConstTransform(in_column="target", value=1)]
>>> pipeline = Pipeline(model, transforms=transforms, horizon=3)
>>> pipeline.set_params(**{"model.lag": 3, "transforms.0.value": 2})
Pipeline(model = NaiveModel(lag = 3, ), transforms = [AddConstTransform(in_column = 'target', value = 2, inplace = True, out_column = None, )], horizon = 3, )

to_dict()

Collect all information about etna object in dict.

class PytorchForecastingMixin

Mixin for Pytorch Forecasting models.

fit(ts: etna.datasets.tsdataset.TSDataset)

Fit model.

Parameters

ts (etna.datasets.tsdataset.TSDataset) – TSDataset to fit.

Returns

model