# Código fonte para qiskit.optimization.applications.ising.vehicle_routing

# This code is part of Qiskit.
#
# (C) Copyright IBM 2019, 2021.
#
# obtain a copy of this license in the LICENSE.txt file in the root directory
#
# Any modifications or derivative works of this code must retain this
# copyright notice, and modified files need to carry a notice indicating
# that they have been altered from the originals.

"""
Converts vehicle routing instances into a list of Paulis,
and provides some related routines (extracting a solution,
checking its objective function value).
"""

from typing import Tuple, List
import numpy as np
from qiskit.quantum_info import Pauli

from qiskit.aqua.algorithms import MinimumEigensolverResult
from qiskit.aqua.operators import WeightedPauliOperator

# pylint: disable=invalid-name

[documentos]def get_vehiclerouting_matrices(instance: np.ndarray,
n: int,
K: int) -> Tuple[np.ndarray, np.ndarray, float]:
"""Constructs auxiliary matrices from a vehicle routing instance,
which represent the encoding into a binary quadratic program.
This is used in the construction of the qubit ops and computation
of the solution cost.

Args:
instance: a customers-to-customers distance matrix.
n: the number of customers.
K: the number of vehicles available.

Returns:
a matrix defining the interactions between variables.
a matrix defining the contribution from the individual variables.
the constant offset.
"""
# N = (n - 1) * n
A = np.max(instance) * 100  # A parameter of cost function

# Determine the weights w
instance_vec = instance.reshape(n ** 2)
w_list = [instance_vec[x] for x in range(n ** 2) if instance_vec[x] > 0]
w = np.zeros(n * (n - 1))
for i_i, _ in enumerate(w_list):
w[i_i] = w_list[i_i]

id_n = np.eye(n)
im_n_1 = np.ones([n - 1, n - 1])
iv_n_1 = np.ones(n)
iv_n_1[0] = 0
iv_n = np.ones(n - 1)
neg_iv_n_1 = np.ones(n) - iv_n_1

v = np.zeros([n, n * (n - 1)])
for i_i in range(n):
count = i_i - 1
for j_j in range(n * (n - 1)):

if j_j // (n - 1) == i_i:
count = i_i

if j_j // (n - 1) != i_i and j_j % (n - 1) == count:
v[i_i][j_j] = 1.

v_n = np.sum(v[1:], axis=0)

# Q defines the interactions between variables
Q = A * (np.kron(id_n, im_n_1) + np.dot(v.T, v))

# g defines the contribution from the individual variables
g = w - 2 * A * (np.kron(iv_n_1, iv_n) + v_n.T) - \
2 * A * K * (np.kron(neg_iv_n_1, iv_n) + v[0].T)

# c is the constant offset
c = 2 * A * (n - 1) + 2 * A * (K ** 2)

return Q, g, c

[documentos]def get_vehiclerouting_cost(instance: np.ndarray, n: int, K: int, x_sol: np.ndarray) -> float:
"""Computes the cost of a solution to an instance of a vehicle routing problem.

Args:
instance: a customers-to-customers distance matrix.
n: the number of customers.
K: the number of vehicles available.
x_sol: a solution, i.e., a path, in its binary representation.

Returns:
objective function value.
"""
(Q, g, c) = get_vehiclerouting_matrices(instance, n, K)

def fun(x):
return np.dot(np.around(x), np.dot(Q, np.around(x))) + np.dot(g, np.around(x)) + c

cost = fun(x_sol)
return cost

[documentos]def get_operator(instance: np.ndarray, n: int, K: int) -> WeightedPauliOperator:
"""Converts an instance of a vehicle routing problem into a list of Paulis.

Args:
instance: a customers-to-customers distance matrix.
n: the number of customers.
K: the number of vehicles available.

Returns:
operator for the Hamiltonian.
"""
N = (n - 1) * n
(Q, g__, c) = get_vehiclerouting_matrices(instance, n, K)

# Defining the new matrices in the Z-basis
i_v = np.ones(N)
q_z = (Q / 4)
g_z = (-g__ / 2 - np.dot(i_v, Q / 4) - np.dot(Q / 4, i_v))
c_z = (c + np.dot(g__ / 2, i_v) + np.dot(i_v, np.dot(Q / 4, i_v)))

c_z = c_z + np.trace(q_z)
q_z = q_z - np.diag(np.diag(q_z))

# Getting the Hamiltonian in the form of a list of Pauli terms

pauli_list = []
for i in range(N):
if g_z[i] != 0:
w_p = np.zeros(N)
v_p = np.zeros(N)
v_p[i] = 1
pauli_list.append((g_z[i], Pauli((v_p, w_p))))
for i in range(N):
for j in range(i):
if q_z[i, j] != 0:
w_p = np.zeros(N)
v_p = np.zeros(N)
v_p[i] = 1
v_p[j] = 1
pauli_list.append((2 * q_z[i, j], Pauli((v_p, w_p))))

pauli_list.append((c_z, Pauli((np.zeros(N), np.zeros(N)))))
return WeightedPauliOperator(paulis=pauli_list)

[documentos]def get_vehiclerouting_solution(instance: np.ndarray,
n: int,
K: int,
result: MinimumEigensolverResult) -> List[int]:
"""Tries to obtain a feasible solution (in vector form) of an instance
of vehicle routing from the results dictionary.

Args:
instance: a customers-to-customers distance matrix.
n: the number of customers.
K: the number of vehicles available.
result: a result obtained by QAOA.run or VQE.run.

Returns:
a solution, i.e., a path, in its binary representation.

#TODO: support statevector simulation, results should be a statevector or counts format, not
a result from algorithm run
"""
del instance, K  # unused
v = result.eigenstate
N = (n - 1) * n

index_value = [x for x in range(len(v)) if v[x] == max(v)][0]
string_value = "{0:b}".format(index_value)

while len(string_value) < N:
string_value = '0' + string_value

x_sol = list()
for elements in string_value:
if elements == '0':
x_sol.append(0)
else:
x_sol.append(1)

x_sol = np.flip(x_sol, axis=0)

return x_sol