motifs2.py

"""
This file implements the efficient algorithm for motif discovery in hypergraphs.
"""

from hypergraph import hypergraph
from utils import *
from loaders import *

def motifs_order_3(edges, TOT):
    N = 3
    full, visited = motifs_ho_full(edges, N, TOT)
    standard = motifs_standard(edges, N, TOT, visited)

    res = []
    for i in range(len(full)):
        res.append((full[i][0], max(full[i][1], standard[i][1])))

    return res

def motifs_order_4(edges, TOT):
    N = 4
    full, visited = motifs_ho_full(edges, N, TOT)
    not_full, visited = motifs_ho_not_full(edges, N, TOT, visited)
    standard = motifs_standard(edges, N, TOT, visited)

    res = []
    for i in range(len(full)):
        res.append((full[i][0], max([full[i][1], not_full[i][1], standard[i][1]])))

    return res

N = 3

edges = load_high_school(N)

output = {}

if N == 3:
    output['motifs'] = motifs_order_3(edges, -1)
elif N == 4:
    output['motifs'] = motifs_order_4(edges, -1)

print(output['motifs'])

STEPS = len(edges)*10
ROUNDS = 10

results = []

for i in range(ROUNDS):
    e1 = hypergraph(edges)
    e1.MH(label='stub', n_steps=STEPS)
    if N == 3:
        m1 = motifs_order_3(e1.C, i)
    elif N == 4:
        m1 = motifs_order_4(e1.C, i)
    results.append(m1)

output['config_model'] = results

delta = diff_sum(output['motifs'], output['config_model'])
norm_delta = norm_vector(delta)

print(norm_delta)