discrete_optimization.generic_tools.robustness package

Submodules

discrete_optimization.generic_tools.robustness.robustness_tool module

class discrete_optimization.generic_tools.robustness.robustness_tool.RobustnessTool(base_instance: RcpspProblem, all_instances: list[RcpspProblem], train_instance: list[RcpspProblem] | None = None, test_instance: list[RcpspProblem] | None = None, proportion_train: float = 0.8)[source]

Bases: object

get_models(aggreg_apriori: bool = True, aggreg_aposteriori: bool = True) → list[RcpspProblem][source]
get_statistics_df(results: ndarray[tuple[Any, ...], dtype[float64]]) → DataFrame[source]

Computes aggregate statistics for each method. Returns a Pandas DataFrame with Mean, Std, Min, Max, and Feasibility %.

plot(results: ndarray[tuple[Any, ...], dtype[float64]], image_tag: str = '') → None[source]

Original plot method (Histograms)

plot_boxplots(results: ndarray[tuple[Any, ...], dtype[float64]], image_tag: str = '') → None[source]

Plots Boxplots for better comparison of distributions and outliers.

solve_and_retrieve(solve_models_function: Callable[[RcpspProblem], ResultStorage], apriori: bool = True, aposteriori: bool = True, nb_process: int = 8) → ndarray[tuple[Any, ...], dtype[float64]][source]
visualize_scenarios(method_tag: str, nb_scenarios: int = 5)[source]

Visualizes the Gantt chart of the solution found by ‘method_tag’ simulated on ‘nb_scenarios’ random test instances.

This uses the native RcpspSolution evaluation mechanism (no Executor).

discrete_optimization.generic_tools.robustness.robustness_tool.solve_model(model: Problem, postpro: bool = True, nb_iteration: int = 500) → ResultStorage[source]

Module contents