simopt.experimental
Experimental framework for running experiments.
Module Contents
- class simopt.experimental.ProblemConfig(name: str, rename: str | None = None, fixed_factors: dict | None = None, model_fixed_factors: dict | None = None)
Class to hold problem config information.
Initialize the Problem with name, rename, and problem/model fixed factors.
- name
- rename
- fixed_factors
- model_fixed_factors
- class simopt.experimental.SolverConfig(name: str, rename: str | None = None, fixed_factors: dict | None = None)
Class to hold solver config information.
Initialize the Solver with name, rename, and solver fixed factors.
- name
- rename
- fixed_factors
- simopt.experimental.run_experiment(problems: list[dict], solvers: list[dict], num_macroreps: int, num_postreps: int, num_postnorms: int) list[list[simopt.experiment_base.ProblemSolver]]
Run an experiment using the provided configurations.
- Parameters:
problems – List of ProblemConfig instances.
solvers – List of SolverConfig instances.
num_macroreps – Number of macroreplications.
num_postreps – Number of post-replications.
num_postnorms – Number of post-normalizations.
- Returns:
A list of lists containing ProblemSolver instances, grouped by problem.
- Return type:
List[list[ProblemSolver]]
- simopt.experimental.group_experiments_by_solver(all_experiments: list[list[simopt.experiment_base.ProblemSolver]]) list[list[simopt.experiment_base.ProblemSolver]]
Group experiments by solver name.