experiment_runner

helper module for running experiments and keeping track of which model has been initialized with which policy.

class ema_workbench.em_framework.experiment_runner.ExperimentRunner(msis)

Helper class for running the experiments

This class contains the logic for initializing models properly, running the experiment, getting the results, and cleaning up afterwards.

Parameters:

  • msis (dict)

  • model_kwargs (dict)

msi_initializiation

keeps track of which model is initialized with which policy.

Type:

dict

msis

models indexed by name

Type:

dict

model_kwargs

keyword arguments for model_init

Type:

dict

run_experiment(experiment)

The logic for running a single experiment. This code makes sure that model(s) are initialized correctly.

Parameters:

experiment (Case instance)

Returns:

  • experiment_id (int)

  • case (dict)

  • policy (str)

  • model_name (str)

  • result (dict)

Raises:

  • EMAError – if the model instance raises an EMA error, these are reraised.

  • Exception – Catch all for all other exceptions being raised by the model. These are reraised.