uneven edge length scaling in networkBFEalloc

[JenkeScheen]

I'm trying to generate a diffnet in spring_layout using a 11x11 symmetrical sij with np.inf on diagonal:
sij_sym.npy.zip

%matplotlib inline
import matplotlib
from matplotlib import pyplot as plt
matplotlib.rcParams['font.size'] = 20
import networkx as nx
import numpy as np
from cvxopt import matrix
import diffnet as dn
import graph as gph
from netbfe import *

sij = matrix(np.sqrt(np.load("sij_sym.npy")))

#insert delta with stdev=1 for first s_ii:
delta = sij.shape[0] * [np.inf]
fict_delta = [1] + delta[1:]

#minimal graph generation:
n = networkBFEalloc( sij, N=1000, delta=fict_delta)
G = gph.diffnet_to_graph( n)
gph.draw_diffnet_graph( G, widthscale=1e-1, nodescale=5)

however this returns a graph where the origin node is distanced, skewing the network:

[forcefield]

Do you have the same problem when the diagonal elements of sij are not np.inf?

[JenkeScheen]

looks like it, yes. This matrix:

produces a similarly skewed network with the same code:

[forcefield]

I pushed a new graph.py that might alleviate this problem. Please see issues/001 for the results.