Source code for quantum_launcher.problems.problem_initialization.maxcut

 1"""  This module contains the MaxCut class."""
 2from typing import Optional
 3import networkx as nx
 4import matplotlib.pyplot as plt
 5
 6from quantum_launcher.base import Problem
 7
 8

[docs]
 9class MaxCut(Problem):
10    """
11    Class for MaxCut Problem.
12
13    This class represents MaxCut Problem which is a combinatorial optimization problem that involves partitioning the
14    vertices of a graph into two sets such that the number of edges between the two sets is maximized. The class contains
15    an instance of the problem, so it can be passed into Quantum Launcher.
16
17    Attributes:
18        instance (nx.Graph): The graph instance representing the problem.
19
20    """
21
22    def __init__(self, instance: nx.Graph, instance_name='unnamed'):
23        super().__init__(instance, instance_name)
24
25    @property
26    def setup(self) -> dict:
27        return {
28            'instance_name': self.instance_name
29        }
30

[docs]
31    @staticmethod
32    def from_preset(instance_name: str) -> "MaxCut":
33        match instance_name:
34            case 'default':
35                node_list = list(range(6))
36                edge_list = [(0, 1), (0, 2), (0, 5), (1, 3), (1, 4),
37                             (2, 4), (2, 5), (3, 4), (3, 5)]
38        graph = nx.Graph()
39        graph.add_nodes_from(node_list)
40        graph.add_edges_from(edge_list)
41        return MaxCut(graph, instance_name=instance_name)

42
43    def _get_path(self) -> str:
44        return f'{self.name}@{self.instance_name}'
45

[docs]
46    def visualize(self, bitstring: Optional[str] = None):
47        pos = nx.spring_layout(self.instance)
48        plt.figure(figsize=(8, 6))
49        if bitstring is None:
50            cmap = 'skyblue'
51        else:
52            cmap = ['crimson' if bit == '1' else 'skyblue' for bit in bitstring]
53        nx.draw(self.instance, pos, with_labels=True, node_color=cmap,
54                node_size=500, edge_color='gray', font_size=10, font_weight='bold')
55        plt.title("Max-Cut Problem Instance Visualization")
56        plt.show()

57

[docs]
58    @staticmethod
59    def generate_maxcut_instance(num_vertices: int, edge_probability: float) -> "MaxCut":
60        graph = nx.gnp_random_graph(num_vertices, edge_probability)
61        return MaxCut(graph, instance_name='Generated')