Dimension.py

import networkx as nx
import matplotlib.pyplot as plt
import math
import numpy as np


#G = nx.erdos_renyi_graph(100, 0.1) test
l_b_max = nx.diameter(G)


def G_prime(G, l_b):
    G_prime = nx.Graph()
    N_prime = len(G.nodes)
    G_prime.add_nodes_from(G)
    E_prime = []
    for n1 in range(N_prime):
        for n2 in range(N_prime):
            try:
                if nx.shortest_path_length(G, n1, n2) > l_b:
                    G_prime.add_edge(n1, n2)
            except nx.NetworkXNoPath:
                return 0
    return G_prime


def Greedy_coloring(G):
    d = nx.coloring.greedy_color(G, strategy="random_sequential")
    return d


def Dimension(G, l_b_max):
    N_bs = np.zeros(l_b_max)
    l_bs = np.zeros(l_b_max)
    for l_b in range(1, l_b_max + 1, 1):
        G1 = G_prime(G, l_b)
        d = Greedy_coloring(G1)
        nx.draw(G, nodes_color=color_maps, with_labels=True)
        plt.show()
        N_b = len(set(d.values()))
        N_bs[l_b - 1] = N_b
        l_bs[l_b - 1] = l_b
    print(N_bs)
    print(l_bs)
    N_bs_log = np.log(N_bs)
    l_bs_log = np.log(l_bs)
    slope, intercept = np.polyfit(l_bs_log, N_bs_log, 1)

    dim = -slope
    print(dim)