SvrHyperparams

svr_model.SvrHyperparams · inherits PydanticBaseModel

Defines the hyperparameters for the SvrModel.

Attributes

Attribute	Type	Default	Description
kernel	Literal["linear", "poly", "rbf", "sigmoid"]	-	SVR kernel type
C	float	-	Regularization parameter
epsilon	float	0.1	Epsilon parameter for epsilon-SVR
gamma	Literal["scale", "auto"]	"scale"	Kernel coefficient for 'rbf', 'poly' and 'sigmoid'

SvrModel

svr_model.SvrModel · inherits BaseModel

Initialize Support Vector Regression (SVR) model with model-specific parameters. Uses direct multi-output strategy via sklearn's MultiOutputRegressor.

__init__(self, params: Dict[str, Any], settings: Dict[str, Any])

Initialize Support Vector Regression (SVR) model with model-specific parameters. Uses direct multi-output strategy via sklearn's MultiOutputRegressor.

Parameter	Type	Default	Description
params	Dict[str, Any]	-	(undocumented)
settings	Dict[str, Any]	-	(undocumented)

train(self, y_context: np.ndarray, y_target: np.ndarray, timestamps_context: np.ndarray, timestamps_target: np.ndarray, x_context: Optional[np.ndarray] = None, x_target: Optional[np.ndarray] = None, **kwargs: dict) -> SvrModel

Train the SVR model for direct multi-output forecasting using MultiOutputRegressor.

Parameter	Type	Default	Description
y_context	np.ndarray	-	(undocumented)
y_target	np.ndarray	-	(undocumented)
timestamps_context	np.ndarray	-	(undocumented)
timestamps_target	np.ndarray	-	(undocumented)
x_context	Optional[np.ndarray]	None	(undocumented)
x_target	Optional[np.ndarray]	None	(undocumented)
**kwargs	dict	-	(undocumented)

Returns: SvrModel (The trained model instance.)

predict(self, y_context: np.ndarray, timestamps_context: np.ndarray, timestamps_target: np.ndarray, x_context: Optional[np.ndarray] = None, x_target: Optional[np.ndarray] = None, **kwargs: dict)

Autoregressive rolling prediction for MultiOutputRegressor SVR. Predicts the entire length of y_target by repeatedly using its own predictions as context.

Parameter	Type	Default	Description
y_context	np.ndarray	-	(undocumented)
timestamps_context	np.ndarray	-	(undocumented)
timestamps_target	np.ndarray	-	(undocumented)
x_context	Optional[np.ndarray]	None	(undocumented)
x_target	Optional[np.ndarray]	None	(undocumented)
**kwargs	dict	-	(undocumented)

Returns: np.ndarray (The predictions, shape (total_steps, num_targets).)

Raises: ValueError: If the model is not trained yet. ValueError: If y_context is shorter than the effective lookback window.