Source code for objective_weighting.mcda_methods.mcda_method

from abc import ABC
import numpy as np

class MCDA_method(ABC):

    def __call__(self, matrix, weights, types):
        """
        Score alternatives from decision matrix `matrix` using criteria weights `weights` and 
        criteria types `types`

        Parameters
        ----------
            matrix : ndarray
                decision matrix with performance values for m alternatives in rows and n criteria 
                in columns
            weights : ndarray
                matrix with criteria weights vectors with number of columns equal to 
                number of columns n of `matrix`
            types : ndarray
                vector with criteria types containing values of 1 for profit criteria and -1 for 
                cost criteria with size equal to number of columns n of `matrix`
        """
        pass

    @staticmethod
    def _verify_input_data(matrix, weights, types):
        m, n = matrix.shape
        # if weights are vector with one dimension
        if len(weights.shape) == 1:
            if len(weights) != n:
                raise ValueError('The size of the weight vector must be the same as the number of criteria')
        # if weights are two-dimensional matrix containing many weight vectors in rows
        elif len(weights.shape) == 2:
            if weights.shape[1] != n:
                raise ValueError('The number of columns of matrix with weight vectors must be the same as the number of criteria')
        if len(types) != n:
            raise ValueError('The size of the types vector must be the same as the number of criteria')
        check_types = np.all((types == 1) | (types == -1))
        if check_types == False:
            raise ValueError('Criteria types can only have a value of 1 for profits and -1 for costs')