gs = GenerationStrategy(
    steps=[

        # 2. Bayesian optimization step (requires data obtained from previous phase and learns
        # from all data available at the time of each new candidate generation call)
        GenerationStep(
            model=Models.GPEI,
            num_trials=-1,  # No limitation on how many trials should be produced from this step
            max_parallelism=1,  # Parallelism limit for this step, often lower than for Sobol
            # More on parallelism vs. required samples in BayesOpt:
            # https://ax.dev/docs/bayesopt.html#tradeoff-between-parallelism-and-total-number-of-trials
        ),
    ]
)

ax_client.create_experiment(
    name="test",
    parameters=[
        {
            "name": "x",
            "type": "choice",
            "value_type": "float",
            "values": list(np.arange(8.0, 15.0, 0.25)),  # Convertissez les éléments en type float natif.
            "is_ordered": True,  # Indique que les valeurs sont ordonnées.
        },
    ],
    choose_generation_strategy_kwargs={"max_parallelism_override": 1},
    objective_name="dsp",
    minimize=True,  # Optional, defaults to False.
    outcome_constraints=["alpha >= 0",
                         "beta >= 0",
                         "gamma >= 0",
                         "delta >= 0",
                         "epsilon >= 0",
                         "iota >= 0",
                         "kappa >= 0",
                        ],  # Optional.
)