Ejemplos de save_bootstrap_mean_evaluation_collection en Python

Ejemplo n.º 1

Archivo: run_eval_normal_dualcontrol_AE.py Proyecto: liuchbryan/experiment_design_evaluation

                               mu_C3=np.random.uniform(-10, 10),
                               mu_Iphi=np.random.uniform(-10, 10),
                               mu_Ipsi=np.random.uniform(-10, 10),
                               sigma_sq_C0=np.random.uniform(1, 10),
                               sigma_sq_C1=np.random.uniform(1, 10),
                               sigma_sq_I1=np.random.uniform(1, 10),
                               sigma_sq_C2=np.random.uniform(1, 10),
                               sigma_sq_I2=np.random.uniform(1, 10),
                               sigma_sq_C3=np.random.uniform(1, 10),
                               sigma_sq_Iphi=np.random.uniform(1, 10),
                               sigma_sq_Ipsi=np.random.uniform(1, 10),
                               n_0=int(50 * 10**np.random.uniform(0, 2.5)),
                               n_1=int(50 * 10**np.random.uniform(0, 2.5)),
                               n_2=int(50 * 10**np.random.uniform(0, 2.5)),
                               n_3=int(50 * 10**np.random.uniform(0, 2.5)),
                               alpha=0.05,
                               pi_min=0.8))

    design4_actual_effect_evaluation = (EDActualEffectEvaluation(
        design4,
        n_init_samples=args.num_init_samples,
        n_bootstrap_mean_samples=args.num_bootstrap_samples))
    design4_actual_effect_evaluation.run(verbose=False)
    design4_actual_effect_evaluations.append(design4_actual_effect_evaluation)

save_bootstrap_mean_evaluation_collection(
    design4_actual_effect_evaluations,
    in_dir=args.output_dir,
    expt_design_name="normal_dualcontrol",
    quantity_name="AE")

Ejemplo n.º 2

Archivo: run_eval_normal_allsamples_AE.py Proyecto: liuchbryan/experiment_design_evaluation

            mu_I1=np.random.uniform(-10, 10),
            mu_C2=np.random.uniform(-10, 10),
            mu_I2=np.random.uniform(-10, 10),
            mu_C3=np.random.uniform(-10, 10),
            mu_Iphi=np.random.uniform(-10, 10),
            mu_Ipsi=np.random.uniform(-10, 10),
            sigma_sq_C0=np.random.uniform(1, 10),
            sigma_sq_C1=np.random.uniform(1, 10),
            sigma_sq_I1=np.random.uniform(1, 10),
            sigma_sq_C2=np.random.uniform(1, 10),
            sigma_sq_I2=np.random.uniform(1, 10),
            sigma_sq_C3=np.random.uniform(1, 10),
            sigma_sq_Iphi=np.random.uniform(1, 10),
            sigma_sq_Ipsi=np.random.uniform(1, 10),
            n_0=int(50 * 10**np.random.uniform(0, 2.5)),
            n_1=int(50 * 10**np.random.uniform(0, 2.5)),
            n_2=int(50 * 10**np.random.uniform(0, 2.5)),
            n_3=int(50 * 10**np.random.uniform(0, 2.5)),
            alpha=0.05, pi_min=0.8
        )
    )

    design2_actual_effect_evaluation = (
        EDActualEffectEvaluation(design2, n_init_samples=args.num_init_samples,
                                 n_bootstrap_mean_samples=args.num_bootstrap_samples))
    design2_actual_effect_evaluation.run(verbose=False)
    design2_actual_effect_evaluations.append(design2_actual_effect_evaluation)

save_bootstrap_mean_evaluation_collection(
    design2_actual_effect_evaluations, in_dir=args.output_dir, expt_design_name="normal_allsample", quantity_name="AE")

Ejemplo n.º 3

                             mu_I2=np.random.uniform(-10, 10),
                             mu_C3=np.random.uniform(-10, 10),
                             mu_Iphi=np.random.uniform(-10, 10),
                             mu_Ipsi=np.random.uniform(-10, 10),
                             sigma_sq_C0=np.random.uniform(1, 10),
                             sigma_sq_C1=np.random.uniform(1, 10),
                             sigma_sq_I1=np.random.uniform(1, 10),
                             sigma_sq_C2=np.random.uniform(1, 10),
                             sigma_sq_I2=np.random.uniform(1, 10),
                             sigma_sq_C3=np.random.uniform(1, 10),
                             sigma_sq_Iphi=np.random.uniform(1, 10),
                             sigma_sq_Ipsi=np.random.uniform(1, 10),
                             n_0=int(50 * 10**np.random.uniform(0, 2.5)),
                             n_1=int(50 * 10**np.random.uniform(0, 2.5)),
                             n_2=int(50 * 10**np.random.uniform(0, 2.5)),
                             n_3=int(50 * 10**np.random.uniform(0, 2.5)),
                             alpha=0.05,
                             pi_min=0.8))

    design2_mde_size_evaluation = (EDMDESizeEvaluation(
        design2,
        n_init_samples=args.num_init_samples,
        n_bootstrap_mean_samples=args.num_bootstrap_samples))
    design2_mde_size_evaluation.run(verbose=True)
    design2_mde_size_evaluations.append(design2_mde_size_evaluation)

save_bootstrap_mean_evaluation_collection(design2_mde_size_evaluations,
                                          in_dir=args.output_dir,
                                          expt_design_name="normal_allsample",
                                          quantity_name="MDES")

Ejemplo n.º 4

Archivo: run_eval_normal_intersectiononly_MDES.py Proyecto: liuchbryan/experiment_design_evaluation

        mu_C3=np.random.uniform(-10, 10),
        mu_Iphi=np.random.uniform(-10, 10),
        mu_Ipsi=np.random.uniform(-10, 10),
        sigma_sq_C0=np.random.uniform(1, 10),
        sigma_sq_C1=np.random.uniform(1, 10),
        sigma_sq_I1=np.random.uniform(1, 10),
        sigma_sq_C2=np.random.uniform(1, 10),
        sigma_sq_I2=np.random.uniform(1, 10),
        sigma_sq_C3=np.random.uniform(1, 10),
        sigma_sq_Iphi=np.random.uniform(1, 10),
        sigma_sq_Ipsi=np.random.uniform(1, 10),
        n_0=int(50 * 10**np.random.uniform(0, 2.5)),
        n_1=int(50 * 10**np.random.uniform(0, 2.5)),
        n_2=int(50 * 10**np.random.uniform(0, 2.5)),
        n_3=int(50 * 10**np.random.uniform(0, 2.5)),
        alpha=0.05,
        pi_min=0.8))

    design1_mde_size_evaluation = (EDMDESizeEvaluation(
        design1,
        n_init_samples=args.num_init_samples,
        n_bootstrap_mean_samples=args.num_bootstrap_samples))
    design1_mde_size_evaluation.run(verbose=True)
    design1_mde_size_evaluations.append(design1_mde_size_evaluation)

save_bootstrap_mean_evaluation_collection(
    design1_mde_size_evaluations,
    in_dir=args.output_dir,
    expt_design_name="normal_intersectiononly",
    quantity_name="MDES")

Ejemplo n.º 5

        mu_C3=np.random.uniform(-10, 10),
        mu_Iphi=np.random.uniform(-10, 10),
        mu_Ipsi=np.random.uniform(-10, 10),
        sigma_sq_C0=np.random.uniform(1, 10),
        sigma_sq_C1=np.random.uniform(1, 10),
        sigma_sq_I1=np.random.uniform(1, 10),
        sigma_sq_C2=np.random.uniform(1, 10),
        sigma_sq_I2=np.random.uniform(1, 10),
        sigma_sq_C3=np.random.uniform(1, 10),
        sigma_sq_Iphi=np.random.uniform(1, 10),
        sigma_sq_Ipsi=np.random.uniform(1, 10),
        n_0=int(50 * 10**np.random.uniform(0, 2.5)),
        n_1=int(50 * 10**np.random.uniform(0, 2.5)),
        n_2=int(50 * 10**np.random.uniform(0, 2.5)),
        n_3=int(50 * 10**np.random.uniform(0, 2.5)),
        alpha=0.05,
        pi_min=0.8))

    design1_actual_effect_evaluation = (EDActualEffectEvaluation(
        design1,
        n_init_samples=args.num_init_samples,
        n_bootstrap_mean_samples=args.num_bootstrap_samples))
    design1_actual_effect_evaluation.run(verbose=False)
    design1_actual_effect_evaluations.append(design1_actual_effect_evaluation)

save_bootstrap_mean_evaluation_collection(
    design1_actual_effect_evaluations,
    in_dir=args.output_dir,
    expt_design_name="normal_intersectiononly",
    quantity_name="AE")

Ejemplo n.º 6

                               mu_C3=np.random.uniform(-10, 10),
                               mu_Iphi=np.random.uniform(-10, 10),
                               mu_Ipsi=np.random.uniform(-10, 10),
                               sigma_sq_C0=np.random.uniform(1, 10),
                               sigma_sq_C1=np.random.uniform(1, 10),
                               sigma_sq_I1=np.random.uniform(1, 10),
                               sigma_sq_C2=np.random.uniform(1, 10),
                               sigma_sq_I2=np.random.uniform(1, 10),
                               sigma_sq_C3=np.random.uniform(1, 10),
                               sigma_sq_Iphi=np.random.uniform(1, 10),
                               sigma_sq_Ipsi=np.random.uniform(1, 10),
                               n_0=int(50 * 10**np.random.uniform(0, 2.5)),
                               n_1=int(50 * 10**np.random.uniform(0, 2.5)),
                               n_2=int(50 * 10**np.random.uniform(0, 2.5)),
                               n_3=int(50 * 10**np.random.uniform(0, 2.5)),
                               alpha=0.05,
                               pi_min=0.8))

    design4_mde_size_evaluation = (EDMDESizeEvaluation(
        design4,
        n_init_samples=args.num_init_samples,
        n_bootstrap_mean_samples=args.num_bootstrap_samples))
    design4_mde_size_evaluation.run(verbose=True)
    design4_mde_size_evaluations.append(design4_mde_size_evaluation)

save_bootstrap_mean_evaluation_collection(
    design4_mde_size_evaluations,
    in_dir=args.output_dir,
    expt_design_name="normal_dualcontrol",
    quantity_name="MDES")