I have been trying to show a graph and I want my edges width to be somewhat related with their weight, meaning tinny when the weight is smaller and larger when the weight is bigger. It makes all edges with the exactly same width. The method amizade is a method that returns the weight. I have imported csv, networkx and mathplotlib.pyplot. This is my code so far:

def nx_teste(graf, show_plot=False):
    nx_g = nx.Graph()  
    for i in graf.edges(): 
        # print(i)
        # print(i._pessoa_1)
        # print(i._pessoa_2)
        nx_g.add_edge(i._pessoa_1._pessoa, i._pessoa_2._pessoa,
                      weight=i._amizade) 
        print(i._pessoa_1, "->", i._pessoa_2, "=", i._amizade)

    if show_plot:
        desenho = nx.spring_layout(nx_g)  
        nx.draw_networkx_nodes(nx_g, desenho, node_size=20,
                               node_color="#32CD32", edgecolors="#8DEEEE")   
        nx.draw_networkx_labels(nx_g, desenho, font_size=7)  
        for f in graf.edges():
            nx.draw_networkx_edges(nx_g, desenho, width=f.amizade()**0.01)  
        nx.draw_networkx_edge_labels(nx_g, desenho, font_size=6)  
        plt.axis("off")  
        plt.show()   

Thank you for your help!

CodePudding user response：

In your code you just need to pass the weights of each edge to width parameter in draw() function.

You can reference the below code. As your code is not reproducible so I have created a random graph

import networkx as nx
import random
G=nx.gnm_random_graph(5,10,seed=42) #creating random graph with 5 nodes and 10 edges
#assigning random weights to each edge
for (u, v) in G.edges():
    G.edges[u,v]['weight'] = random.randint(0,5)

#weight of each of the edge
weights = nx.get_edge_attributes(G,'weight').values()
#drawing the graph
nx.draw(G,pos=nx.spring_layout(G),width=list(weights))

The graph looks something like this: