Finding the optimum value of all Decision Variables in a MixedVariable problem with Pymoo

I've successfully built my model and it runs and plots a Pareto front. However, I don't know how to store X* (the optimum value of ALL the decision variables that resulted in the optimum F).

My model is MixedVariable and I have both Integer and Real variables as decision variables, but I don't know how to print or store the value of every single one of them.

I read the documentation on the Pymoo website but I couldn't materialize them, that is, when I do as documentation instructions, it only prints one value per each defined category of variables, instead of the values of all the decision variables belonging to that category (by category I mean set of the variables with the same type and range). I would place a down-scaled example of my script in the snippet below:

import numpy as np
import pandas as pd
from pymoo.core.problem import ElementwiseProblem
from pymoo.core.variable import Real, Integer

class MyProblem(ElementwiseProblem):
    def __init__(self, I_, J_, d__ij, d__ji):
        self.I = I_
        self.J = J_
        self.d_ij = d__ij
        self.d_ji = d__ji
        J, I = self.J, self.I

        n_obj = 2
        n_ieq_constr = I * J
        n_eq_constr = J + 3*I
        Vars = {
            "y": Integer(bounds=(0, 100)),
            "z": Integer(bounds=(0, 1)),
            "w": Real(bounds=(0, 100)),
            "x": Real(bounds=(0, 1))
        }
        super().__init__(vars=Vars,
                         n_obj=n_obj,
                         n_ieq_constr=n_ieq_constr,
                         n_eq_constr=n_eq_constr)

    def _evaluate(self, X, out, *args, **kwargs):
        J, I = self.J, self.I
        x = []
        for i in range(J*I):
            xi = X["x"]
            x.append(xi)
        y = []
        for i in range(2*I + J):
            yi = X["y"]
            y.append(yi)
        z = []
        for i in range(4*J + 1):
            zi = X["z"]
            z.append(zi)
        w = []
        for i in range(I + J):
            wi = X["w"]
            w.append(wi)

        # Decision Variables:
        Pr_i = np.array(w[:I]).reshape(I, 1)
        IB_j = np.array(w[I:I+J]).reshape(J, 1)
        ...

        CRM_ji = np.array(x[:J*I]).reshape(J, I)
        ...

        R_ij = np.array(y[: I*J]).reshape(I, J)
        S_ij = np.array(y[J*I : J*I+J*I]).reshape(I, J)
        ...

        X_j = np.array(z[:J]).reshape(J, 1)
        X_i = np.array(z[J:J+I]).reshape(I, 1)
        ...

        # Objective functions
        f1 = F(self.d_ji, R_ij, Pr_i, ...)
        f2 = F(self.d_ij, X_j, CRM_ji, ...)

        out["F"] = [f1, f2]
        out["H"] = [h1, ..., hn]
        out["G"] = [g1, ..., gm]

# Load data I, J, d_ij, d_ji
problem = MyProblem(I, J, d_ij, d_ji)

from pymoo.core.mixed import MixedVariableGA
from pymoo.optimize import minimize
from pymoo.algorithms.moo.nsga3 import ReferenceDirectionSurvival
from pymoo.util.ref_dirs import get_reference_directions
ref_dirs = get_reference_directions("das-dennis", 2, n_partitions=12)
algorithm = MixedVariableGA(pop_size=200, survival=ReferenceDirectionSurvival(ref_dirs))

res = minimize(problem,
               algorithm,
               ('n_gen', 15),
               seed=42,
               verbose=True)

print("Best solution found: \nX = %s" % (res.X))
from pymoo.visualization.scatter import Scatter
plot = Scatter()
plot.add(problem.pareto_front(), plot_type="line", color="black", alpha=0.7)
plot.add(res.F, facecolor="red", edgecolor="none")
plot.show()

If I assign res.X to a variable, for example named X, it will be a Numpy array containing pop_size rows of dict objects each containing only four single key-value pairs like this {'y': 0, 'z': 1, 'w': 4.603468972784174, 'x': 0.031364685547833934}; however y, z, w, and x are not supposed to be single floats or integers, but list! as I have determined their type earlier in the _evaluate function.

How can I derive the values for all decision variables?