Surrogate Model
Contains the implementations for the surrogate modelling methods
Implements the supported surrogate modelling methods.
- class calisim.surrogate.implementation.SurrogateModelMethod(calibration_func: Callable, specification: SurrogateModelMethodModel, engine: str = 'sklearn', implementation: CalibrationWorkflowBase | None = None)[source]
The surrogate modelling method class.
- pydantic model calisim.surrogate.implementation.SurrogateModelMethodModel[source]
The surrogate modelling method data model.
- Parameters:
BaseModel (CalibrationModel) – The calibration base model class.
- Config:
arbitrary_types_allowed: bool = True
protected_namespaces: tuple = ()
- Fields:
- field batch_size: int = 1000
The batch size when training the surrogate model
- field flatten_Y: bool = False
Flatten the simulation outputs
- calisim.surrogate.implementation.get_implementations() dict[str, type[CalibrationWorkflowBase]][source]
Get the calibration implementations for surrogate modelling.
- Returns:
- The dictionary of
calibration implementations for surrogate modelling.
- Return type:
Dict[str, type[CalibrationWorkflowBase]]
GPyTorch
Contains the implementations for surrogate modelling methods using GPyTorch
Implements the supported surrogate modelling methods using the GPyTorch library.
- class calisim.surrogate.gpytorch_wrapper.GPyTorchSurrogateModel(calibration_func: Callable, specification: CalibrationModel, task: str)[source]
The GPyTorch surrogate modelling method class.
Scikit-Learn
Contains the implementations for surrogate modelling methods using Scikit-Learn
Implements the supported surrogate modelling methods using the Scikit-Learn library.
- class calisim.surrogate.sklearn_wrapper.SklearnSurrogateModel(calibration_func: Callable, specification: CalibrationModel, task: str)[source]
The Scikit-Learn surrogate modelling method class.