Source code for gemseo_pymoo.post.compromise

# Copyright 2022 Airbus SAS
# Copyright 2021 IRT Saint Exupéry, https://www.irt-saintexupery.com
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU Lesser General Public
# License version 3 as published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
# Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
#
# Contributors:
#    INITIAL AUTHORS - initial API and implementation and/or initial
#                           documentation
#        :author: Gabriel Max DE MENDONÇA ABRANTES
"""Compromise points for multi-criteria decision-making."""

from __future__ import annotations

import logging
from typing import Any

from gemseo.algos.pareto import ParetoFront
from numpy import atleast_2d
from numpy import ndarray
from numpy import vstack
from pymoo.core.decomposition import Decomposition
from pymoo.decomposition.weighted_sum import WeightedSum

from gemseo_pymoo.post.scatter_pareto import ScatterPareto

LOGGER = logging.getLogger(__name__)




class Compromise(ScatterPareto):
    """Scatter plot with pareto front and compromise points.

    See
    `Compromise Programming <https://pymoo.org/mcdm/index.html#Compromise-Programming>`_.
    """

    fig_title = "Compromise Points"

    fig_name_prefix = "compromise"

    def _plot(
        self,
        decomposition: Decomposition | None = None,
        weights: ndarray | None = None,
        plot_extra: bool = True,
        plot_legend: bool = True,
        plot_arrow: bool = False,
        **scalar_options: Any,
    ) -> None:
        """Scatter plot of the pareto front along with the compromise points.

        The compromise points are calculated using a
        `scalarization function <https://pymoo.org/misc/decomposition.html>`_).

        Args:
            decomposition: The instance of the scalarization function to use. If
                ``None``, use a weighted sum.
            weights: The weights for the scalarization function. If None, a normalized
                array is used, e.g. [1./n, 1./n, ..., 1./n] for an optimization problem
                with n-objectives.
            plot_extra: Whether to plot the extra pareto related points,
                i.e. ``utopia``, ``nadir`` and ``anchor`` points.
            plot_legend: Whether to show the legend.
            plot_arrow: Whether to plot arrows connecting the utopia point to
                the compromise points. The arrows are annotated with the ``2-norm`` (
                `Euclidian distance <https://en.wikipedia.org/wiki/Euclidean_distance>`_
                ) of the vector represented by the arrow.
            **scalar_options: The keyword arguments for the scalarization function.

        Raises:
            TypeError: If the scalarization function is not an instance of
                ``Decomposition``.
            ValueError: If the number of weights does not match the number
                 of objectives.
        """
        if decomposition is None:
            decomposition = WeightedSum()
        elif not isinstance(decomposition, Decomposition):
            msg = (
                "The scalarization function must be an instance of "
                "pymoo.core.Decomposition."
            )
            raise TypeError(msg)

        # Objectives.
        n_obj = self.opt_problem.objective.dim

        # Default weights.
        if weights is None:
            weights = [1.0 / n_obj] * n_obj

        # Ensure correct dimension and type.
        weights = atleast_2d(weights).astype(float)

        # Check weight's dimension.
        if weights.shape[1] != n_obj:
            msg = "You must provide exactly one weight for each objective function!"
            raise ValueError(msg)

        # Create Pareto object.
        pareto = ParetoFront.from_optimization_problem(self.opt_problem)

        # Prepare points to plot.
        points = []  # Points' coordinates.
        point_labels = []  # Points' labels.
        for weight in weights:
            # Apply decomposition.
            d_res = decomposition.do(
                pareto.f_optima,
                weight,
                utopian_point=pareto.f_utopia,
                nadir_point=pareto.f_anti_utopia,
                **scalar_options,
            )

            # Best value according to the scalarization function.
            d_min = d_res.min()

            # Index where the minimum value is located (at the pareto front).
            d_idx = d_res.argmin()

            # Point's coordinates.
            points.append(pareto.f_optima[d_idx])

            # Point's label.
            float_format = ".2e" if abs(d_min) > 1e3 else ".2f"
            point_labels.append(f"s({weight}) = {d_min:{float_format}}")

        # Concatenate points to plot.
        points = vstack(points)

        # Extra figure options.
        self.fig_title = f"{self.fig_title}\n(s = {decomposition.__class__.__name__})"

        # Update name's prefix with current decomposition function's name.
        self.fig_name_prefix += f"_{decomposition.__class__.__name__}"

        super()._plot(points, point_labels, plot_extra, plot_legend, plot_arrow)