ExponentialSmoothing

Model

DashAI.back.models.forecasting.ExponentialSmoothing

Forecast from a weighted average that favours recent observations.

Exponential smoothing tracks the level of a series by averaging its past, weighting each observation less the older it is. Optional trend and seasonal components extend that to series which drift or repeat, which together make up the Holt-Winters method.

Where ARIMA models a series through its correlations, this models it through its structure: level, direction, and repeating pattern. That makes it the better first choice when a series visibly has a season, and it is usually easier to configure, since the three components are things you can see in a plot rather than orders you have to search for.

Additive components suit a swing of roughly constant size; multiplicative ones suit a swing that grows as the series does. A seasonal component needs a season length above 1 and at least two full cycles of history.

References

Parameters

trend : string, default=none: Whether the series drifts steadily up or down. 'add' for a trend of roughly constant size, 'mul' for one that grows with the level, 'none' for a series with no direction.
seasonal : string, default=none: Whether a pattern repeats every season. 'add' when the swing is about the same size each cycle, 'mul' when it grows with the level. Needs a season length above 1.
season_length : integer, default=1: How many observations make up one full cycle: 12 for monthly data repeating yearly, 7 for daily data repeating weekly. Only used when a seasonal component is selected.

Methods

predict(self, x: 'DashAIDataset') -> 'np.ndarray'

Defined on ExponentialSmoothing

Forecast forward from the end of the training series.

Parameters

x : DashAIDataset: The rows to forecast, whose dates say how far ahead each one is.

Returns

np.ndarray: One forecast value per requested row.

train(self, x_train: 'DashAIDataset', y_train: 'DashAIDataset', x_validation: 'DashAIDataset' = None, y_validation: 'DashAIDataset' = None) -> 'ExponentialSmoothing'

Defined on ExponentialSmoothing

Fit the smoothing model to the series.

Parameters

x_train : DashAIDataset: The date column. statsmodels is given the values in order and the spacing is assumed regular; the dates are kept so a later partition can be forecast at its own dates.
y_train : DashAIDataset: The series to forecast.
x_validation : DashAIDataset, optional: Unused.
y_validation : DashAIDataset, optional: Unused.

Returns

ExponentialSmoothing: The fitted model.

calculate_metrics(self, split: DashAI.back.core.enums.metrics.SplitEnum = <SplitEnum.VALIDATION: 'validation'>, level: DashAI.back.core.enums.metrics.LevelEnum = <LevelEnum.LAST: 'last'>, log_index: int = None, x_data: 'DashAIDataset' = None, y_data: 'DashAIDataset' = None, fold_index: int = None, inner_fold_index: int = None)

Defined on BaseModel

Calculate and save metrics for a given data split and level.

Parameters

split : SplitEnum: The data split to evaluate (TRAIN, VALIDATION, or TEST). Defaults to SplitEnum.VALIDATION.
level : LevelEnum: The metric granularity level (LAST, TRIAL, STEP, or BATCH). Defaults to LevelEnum.LAST.
log_index : int, optional: Explicit step index for the metric entry. If None, the next step index is computed automatically. Defaults to None.
x_data : DashAIDataset, optional: Input features. If None, the dataset stored in the model for the given split is used. Defaults to None.
y_data : DashAIDataset, optional: Target labels. If None, the labels stored in the model for the given split are used. Defaults to None.

compute_metrics(self, split: DashAI.back.core.enums.metrics.SplitEnum = <SplitEnum.TEST: 'test'>, x_data: 'DashAIDataset' = None, y_data: 'DashAIDataset' = None) -> Dict[str, float]

Defined on BaseModel

Calculate and return metric scores for a given data split.

Parameters

split : SplitEnum: The data split to evaluate (TRAIN, VALIDATION, or TEST). Defaults to SplitEnum.VALIDATION.
x_data : DashAIDataset, optional: Input features. If None, the dataset stored in the model for the given split is used. Defaults to None.
y_data : DashAIDataset, optional: Target labels. If None, the labels stored in the model for the given split are used. Defaults to None.

Returns

Dict[str, float]: A dictionary mapping metric names to their computed scores.

get_credential(self, name: str)

Defined on ConfigObject

Resolve a registered credential component by name.

Parameters

name : str: Credential component class name (e.g. "HuggingFaceCredential").

Returns

BaseCredential: An instance of the requested credential component.

get_metadata(cls) -> Dict[str, Any]

Defined on BaseModel

Get metadata values for the current model.

Returns

Dict[str, Any]: Dictionary containing UI metadata such as the model icon used in the DashAI frontend.

get_schema(cls) -> dict

Defined on ConfigObject

Generates the component related Json Schema.

Returns

dict: Dictionary representing the Json Schema of the component.

load(filename: str) -> Any

Defined on ForecastingModel

Deserialise a model from disk using joblib.

Parameters

filename : str: Path to the file previously written by :meth:save.

Returns

ForecastingModel: The loaded model instance.

predict_prepared(self, features: Any) -> Any

Defined on BaseModel

Predict from data that is already in this model's feature space.

Parameters

features : pandas.DataFrame or numpy.ndarray: Feature matrix as returned by prepare_dataset(..., is_fit=False). No further preparation is applied to it.

Returns

Any: The same kind of output as predict: predicted values for regressors, class probabilities for DashAI classifiers.

predict_proba_prepared(self, features: Any) -> Any

Defined on BaseModel

Return class probabilities for data already in the feature space.

Parameters

features : pandas.DataFrame or numpy.ndarray: Feature matrix as returned by prepare_dataset(..., is_fit=False).

Returns

numpy.ndarray: Array of shape (n_samples, n_classes) with class probabilities.

prepare_dataset(self, dataset: 'DashAIDataset', is_fit: bool = False) -> 'DashAIDataset'

Defined on BaseModel

Hook for model specific preprocessing of input features.

Parameters

dataset : DashAIDataset: The input dataset to preprocess.
is_fit : bool: Whether the call is part of a fitting phase. Defaults to False.

Returns

DashAIDataset: The preprocessed dataset ready to be fed into the model.

prepare_output(self, dataset: 'DashAIDataset', is_fit: bool = False) -> 'DashAIDataset'

Defined on BaseModel

Hook for model-specific preprocessing of output targets.

Parameters

dataset : DashAIDataset: The output dataset (target labels) to preprocess.
is_fit : bool: Whether the call is part of a fitting phase. Defaults to False.

Returns

DashAIDataset: The preprocessed output dataset.

save(self, filename: str) -> None

Defined on ForecastingModel

Serialise the model to disk using joblib.

Parameters

filename : str: Destination file path where the model will be written.

validate_and_transform(self, raw_data: dict) -> dict

Defined on ConfigObject

It takes the data given by the user to initialize the model and returns it with all the objects that the model needs to work.

Parameters

raw_data : dict: A dictionary with the data provided by the user to initialize the model.

Returns

dict: A validated dictionary with the necessary objects.

Compatible with

ForecastingTask

References​

Parameters​

Methods​

Compatible with​

References

Parameters

Methods

Compatible with