QP.py

import numpy as np
from cvxopt import matrix, solvers


class QP(object):

    def __init__(self):
            super().__init__()
            self.LagrangeMultipliers = None

    def QP(self, Q: np.array, c: np.array, A: np.array, b: np.array) -> np.array:
            """Minimize_x 1/2*x^T*Q*x+c^T*x subject to A*x<=b for given matrices Q (symmetric) and A and given vectors c and b. """
            # long live global options, which cannot be set locally
            # sol = solvers.qp(matrix(np.asfarray(Q)), matrix(np.asfarray(c)),
            #                   matrix(np.asfarray(A)), matrix(np.asfarray(b)), options={'show_progress': False})
            
            # sol with np.asarray with dtype float, because np.asfarray deprecated since numpy 2.0
            sol = solvers.qp(matrix(np.asarray(Q, dtype=float)), matrix(np.asarray(c, dtype=float)),
                              matrix(np.asarray(A, dtype=float)), matrix(np.asarray(b, dtype=float)), options={'show_progress': False})
            self.LagrangeMultipliers = np.array(sol['z'])
            return np.array(sol['x']).flatten()


# qp = QP()
# print(qp.QP(np.array([[2.0]]), np.array([[-2]]),
#       np.array([[1.0]]), np.array([0.5])))
# print(qp.QP(np.array([[2.0]]), np.array([[-2]]),
#       np.array([[1.0]]), np.array([1.0])))
# print(qp.QP(np.array([[2.0]]), np.array([[-2]]),
#       np.array([[1.0]]), np.array([1.5])))