Data Model

Simulation

Contains Pydantic data models for the simulation

Several Pydantic data models are defined for various simulation calibration procedures.

pydantic model calisim.data_model.simulation_data_models.BaseModel[source]

Base Pydantic data model.

Parameters:

PydanticBaseModel (PydanticBaseModel) – The Pydantic Base model class.

Config:

  • arbitrary_types_allowed: bool = True

  • protected_namespaces: tuple = ()

class Config[source]
pydantic model calisim.data_model.simulation_data_models.CalibrationModel[source]

The calibration data model.

Parameters:

BaseModel (BaseModel) – The Pydantic Base model class.

Config:

  • arbitrary_types_allowed: bool = True

  • protected_namespaces: tuple = ()

Fields:
field X: ndarray | DataFrame | list | None = None

The simulation input data

field Y: ndarray | DataFrame | list | None = None

The simulation output data

field analyze_kwargs: dict[str, Any] | None = None

The analyze step named arguments

field batched: bool = False

Whether to batch the simulations

field calibration_func_kwargs: dict[str, Any] | None = None

The calibration function named arguments

field experiment_name: str | None = 'default'

The modelling experiment name

field figsize: tuple[int, int] = (12, 12)

The figure size for visualisations

field groups: list[str] | None = None

The list of parameter groups

field lr: float = 0.01

The learning rate of the model

field method: str = ''

The calibration method or algorithm

field method_kwargs: dict[str, Any] | None = None

The calibration method named arguments

field n_chains: int = 1

The number of Markov chains

field n_init: int = 1

The number of initial samples or steps

field n_iterations: int = 1

The number of iterations for sequential calibrators

field n_jobs: int = 1

The number of jobs to run in parallel

field n_out: int = 1

Number of simulation outputs

field n_replicates: int = 1

The number of replicate simulations to run

field n_samples: int = 1

The number of samples to take

field observed_data: ndarray | DataFrame | float | None = None

The empirical or observed data

field outdir: str | None = None

The output directory for modelling results

field output_labels: list[str] | None = None

The list of simulation output names

field parallel_backend: str = ''

The backend engine to run parallel jobs

field parameter_spec: ParameterSpecification | None = None

The parameter specification

field pass_calibration_workflow: bool | str | None = None

Pass the calibration workflow into the calibration function

field random_seed: int | None = None

The random seed for replicability

field storage: str | None = None

The storage backend to persist the calibration results

field test_size: float | int = 0.0

The size of the testing dataset.

field use_shap: bool = False

Whether to use SHAP for feature explanations

field verbose: bool = False

Whether to print calibration messages

field walltime: int = 1

The maximum calibration walltime

pydantic model calisim.data_model.simulation_data_models.DistributionModel[source]

The probability distribution data model.

Parameters:

ParameterModel (ParameterModel) – The simulation parameter data model.

Config:

  • arbitrary_types_allowed: bool = True

  • protected_namespaces: tuple = ()

Fields:
field distribution_args: list | None = None

The distribution positional arguments

field distribution_bounds: list | None = None

The distribution bounds

field distribution_kwargs: dict[str, Any] | None = None

The distribution named arguments

field distribution_name: str = 'uniform'

The distribution name

pydantic model calisim.data_model.simulation_data_models.OrchestrationModel[source]

The orchestration data model.

Parameters:

BaseModel (BaseModel) – The Pydantic Base model class.

Config:

  • arbitrary_types_allowed: bool = True

  • protected_namespaces: tuple = ()

Fields:
field auto_set_cuda_visible_devices: bool = True

Set the CUDA_VISIBLE_DEVICES variable

field cpu: int = 1

The number of CPUs

field entrypoint: str = 'sh'

The job entrypoint

field gpu: int = 0

The number of GPUs

field image: str = ''

The job Docker image

field log_dir: str | None = None

The log directory

field memMB: int = 100

The amount of memory in MB

field name: str = 'default'

The job name

field num_replicas: int = 1

The number of job replicas

field partition: str = 'fast'

The job partition

field scheduler: str = 'local_cwd'

The orchestration scheduler

field time: str = '00:00:59'

The job walltime

field wait_interval: int = 10

The interval for job status updates (seconds)

class calisim.data_model.simulation_data_models.ParameterDataType(value)[source]

An enumeration.

pydantic model calisim.data_model.simulation_data_models.ParameterEstimateModel[source]

The simulation parameter estimate data model.

Parameters:

BaseModel (BaseModel) – The Pydantic Base model class.

Config:

  • arbitrary_types_allowed: bool = True

  • protected_namespaces: tuple = ()

Fields:
field estimate: float [Required]

The estimated parameter value; a point estimate

field name: str [Required]

The parameter name.

field uncertainty: float | None = None

The uncertainty of the parameter estimate

pydantic model calisim.data_model.simulation_data_models.ParameterEstimatesModel[source]

The simulation parameter estimates data model.

Parameters:

BaseModel (BaseModel) – The Pydantic Base model class.

Config:

  • arbitrary_types_allowed: bool = True

  • protected_namespaces: tuple = ()

Fields:
field estimates: list[ParameterEstimateModel] | None = None

The set of parameter estimates

pydantic model calisim.data_model.simulation_data_models.ParameterModel[source]

The simulation parameter data model.

Parameters:

BaseModel (BaseModel) – The Pydantic Base model class.

Config:

  • arbitrary_types_allowed: bool = True

  • protected_namespaces: tuple = ()

Fields:
field data_type: ParameterDataType = ParameterDataType.CONTINUOUS

The parameter data type

field name: str [Required]

The parameter name

field parameter_tags: dict[str, str] | None = None

A collection of metadata tags.

field parameter_value: float | str | int | None = None

The parameter value.

field parameter_values: list[float] | list[str] | list[int] | None = None

The list of parameter values.

pydantic model calisim.data_model.simulation_data_models.ParameterSpecification[source]

The collection of parameters.

Parameters:

BaseModel (BaseModel) – The Pydantic Base model class.

Config:

  • arbitrary_types_allowed: bool = True

  • protected_namespaces: tuple = ()

Fields:
field parameters: list[DistributionModel] | None = None

The parameter specification list