def multi_abp(G, r, lambda1, m, mp, dim, K):
N = len(G.nodes())
mt = np.zeros((N, dim))
for k in range(dim):
y_abp = SBM.SBM_ABP(G, r, lambda1, m, mp)
mt[:, k] = y_abp
k_means = KMeans(n_clusters=K, max_iter=100, precompute_distances=False)
k_means.fit(mt)
y = k_means.labels_
return y
nmi_arry = {}
ccr_arry = {}
ars_arry = {}
# parameter setting
# c_array = [0.2, 0.4, 0.5, 0.6, 0.8, 1.0, 1.5, 2.0] # alpha = clog(n)/n
# c_array = [0.4, 0.6, 0.8, 1.0, 1.5, 2.0, 4.0] # alpha = clog(n)/n
c_array = [0.6, 0.8, 1.0, 1.2, 1.3, 1.5, 2.0, 2.5, 3.0,
4.0] # alpha = clog(n)/n
K_array = [5] # number of communities
N_array = [1000] # number of nodes
lambda_array = [0.9] # B0 = lambda*I + (1-lambda)*ones(1,1)
# scanning through parameters
for c, K, N, lambda_n in itertools.product(c_array, K_array, N_array,
lambda_array):
print 'K:', K, 'N:', N, 'c:', c, 'lambda:', lambda_n
model_sbm1 = SBM.SBM_param_init1(K, N, lambda_n, c)
for rand in range(rand_tests):
strsub1 = 'K' + str(K) + 'N' + str(N) + 'c' + str(c) + 'la' + str(
lambda_n) + 'rd' + str(rand) # for saving results
# simulate graph
G = SBM.SBM_simulate_fast(model_sbm1)
ln, nodeslist = get_label_list(G)
# algo1: proposed deepwalk algorithm
model_w2v = SBM.SBM_learn_deepwalk_1(G, num_paths, length_path,
emb_dim, rw_filename,
emb_filename, winsize)
X = model_w2v[nodeslist]
k_means = KMeans(n_clusters=K,
max_iter=100,
precompute_distances=False)
k_means.fit(X)
# generating multiple graphs for the same parameter setting
rand_tests = 5
# setting storage space for results
nmi_arry = {}
ccr_arry = {}
ars_arry = {}
# parameter setting
c_array = [2] # alpha = clog(n)/n
K_array = [5] # number of communities
N_array = [10000] # number of nodes
lambda_array = [0.9] # B0 = lambda*I + (1-lambda)*ones(1,1)
# scanning through parameters
for c, K, N, lambda_n in itertools.product(c_array, K_array, N_array,
lambda_array):
print 'K:', K, 'N:', N, 'c:', c, 'lambda:', lambda_n
model_sbm1 = SBM.SBM_param_init1(K, N, lambda_n, c)
for rand in range(rand_tests):
strsub1 = 'K' + str(K) + 'N' + str(N) + 'c' + str(c) + 'la' + str(
lambda_n) + 'rd' + str(rand) # for saving results
# simulate graph
G = SBM.SBM_simulate_fast(model_sbm1)
ln, nodeslist = get_label_list(G)
# algo1: proposed deepwalk algorithm
# scanning parameter settings
emb_dimensions = [50]
winsize_arry = [8]
num_path_arry = [5, 10, 15, 20, 25, 30]
length_path_arry = [60]
for length_path, num_paths in itertools.product(
length_path_arry, num_path_arry):
# creating random walks and save in memory