Evolutionary Algorithms

Contains the implementations for the evolutionary algorithm methods

Implements the supported evolutionary algorithm methods.

class calisim.evolutionary.implementation.EvolutionaryMethod(calibration_func: Callable, specification: EvolutionaryMethodModel, engine: str = 'spotpy', implementation: type[CalibrationWorkflowBase] | CalibrationWorkflowBase | None = None)[source]: The evolutionary algorithm method class.

pydantic model calisim.evolutionary.implementation.EvolutionaryMethodModel[source]

The evolutionary algorithm method data model.

Parameters:

BaseModel (CalibrationModel) – The calibration base model class.

Config:

arbitrary_types_allowed: bool = True
protected_namespaces: tuple = ()

Fields:

directions (list[str] | None)
objective (str | None)
operators (dict[str, Any] | None)

field directions: list[str] | None = ['minimize']: The list of objective directions

field objective: str | None = 'rmse': The objective function

field operators: dict[str, Any] | None = None: The dictionary of evolutionary operators

calisim.evolutionary.implementation.get_implementations() → dict[str, str][source]

Get the calibration implementations for evolutionary algorithm.

Returns:

The dictionary: of calibration implementations for evolutionary algorithm.

Return type:

Dict[str, str]

EvoTorch

Contains the implementations for evolutionary algorithms using EvoTorch

Implements the supported evolutionary algorithms using the EvoTorch library.

class calisim.evolutionary.evotorch_wrapper.EvoTorchEvolutionary(calibration_func: Callable, specification: CalibrationModel, task: str)[source]

The EvoTorch evolutionary algorithm method class.

analyze() → None[source]: Analyze the results of the simulation calibration procedure.

execute() → None[source]: Execute the simulation calibration procedure.

specify() → None[source]: Specify the parameters of the model calibration procedure.

SPOTPY

Contains the implementations for evolutionary algorithms using SPOTPY

Implements the supported evolutionary algorithms using the SPOTPY library.

class calisim.evolutionary.spotpy_wrapper.SPOTPYEvolutionary(calibration_func: Callable, specification: CalibrationModel, task: str)[source]

The SPOTPY evolutionary algorithm method class.

analyze() → None[source]: Analyze the results of the simulation calibration procedure.

execute() → None[source]: Execute the simulation calibration procedure.

specify() → None[source]: Specify the parameters of the model calibration procedure.

class calisim.evolutionary.spotpy_wrapper.SPOTSetup(workflow: CalibrationWorkflowBase, objective_function: Callable, evolutionary_name: str)[source]

The SPOTPY calibration setup.

evaluation() → ndarray | DataFrame[source]

Get the observed data.

Returns:: The observed data.
Return type:: np.ndarray | pd.DataFrame

objectivefunction(simulation: ndarray | DataFrame, evaluation: ndarray | DataFrame) → float[source]

Call the objective function on simulated and observed data.

Parameters:

simulation (np.ndarray | pd.DataFrame) – The simulated data.
evaluation (np.ndarray | pd.DataFrame) – The observed data.

Returns:

The objective function results.

Return type:

float

parameters() → ndarray[source]

Generate parameters from the prior specification.

Returns:: The generated parameters.
Return type:: np.ndarray

setup_from_workflow(workflow: CalibrationWorkflowBase) → None[source]

Configure the calibration procedure from the workflow object.

Parameters:: workflow (CalibrationWorkflowBase) – The calibration workflow object.

simulation(X: ndarray) → ndarray[source]

Run the simulation.

Parameters:: X (np.ndarray) – The simulation parameter vector.
Returns:: The simulation results.
Return type:: np.ndarray