n_batch=500,
modelfile=[
'gem/intermediate/enc_model.json',
'gem/intermediate/dec_model.json'
],
weightfile=[
'gem/intermediate/enc_weights.hdf5',
'gem/intermediate/dec_weights.hdf5'
])
G_X = nx.to_numpy_matrix(G)
embedding.learn_embedding(G)
G_X_hat = embedding.get_reconstructed_adj()
rec_norm = np.linalg.norm(G_X - G_X_hat)
print(rec_norm)
import pdb
pdb.set_trace()
# X = embedding.get_embedding()
# import pdb
# pdb.set_trace()
node_colors_arr = [None] * node_colors.shape[0]
for idx in range(node_colors.shape[0]):
node_colors_arr[idx] = np.where(
node_colors[idx, :].toarray() == 1)[1][0]
# MAP, prec_curv, err, err_baseline = gr.evaluateStaticGraphReconstruction(
# G, embedding, X, None
# )
# print('MAP:')
# print(MAP)
viz.plot_embedding2D(G_X, di_graph=G, node_colors=node_colors_arr)
plt.savefig('sdne_sbm_g_x.pdf', bbox_inches='tight')
else:
node_num = self._node_num
adj_mtx_r = np.zeros((node_num, node_num))
for v_i in range(node_num):
for v_j in range(node_num):
if v_i == v_j:
continue
adj_mtx_r[v_i, v_j] = self.get_edge_weight(v_i, v_j)
return adj_mtx_r
if __name__ == '__main__':
# load Zachary's Karate graph
edge_f = 'data/karate.edgelist'
G = graph_util.loadGraphFromEdgeListTxt(edge_f, directed=False)
G = G.to_directed()
res_pre = 'results/testKarate'
graph_util.print_graph_stats(G)
t1 = time()
embedding = LocallyLinearEmbedding(2)
embedding.learn_embedding(graph=G,
edge_f=None,
is_weighted=True,
no_python=True)
print('Graph Factorization:\n\tTraining time: %f' % (time() - t1))
viz.plot_embedding2D(embedding.get_embedding(),
di_graph=G,
node_colors=None)
plt.show()
#---------------------------------------------------------------------------------
print(("\tMAP: {} \t precision curve: {}\n\n\n\n" + '-' * 100).format(
MAP, prec_curv[:5]))
#---------------------------------------------------------------------------------
print("DEBUG::embedding vectors:")
print(embedding.get_embedding())
# cluster vectors (community detection!)
kmeans = KMeans(n_clusters=2,
random_state=0).fit(embedding.get_embedding())
print(kmeans.labels_)
# Visualize using GEM library
X_ = embedding.get_embedding()
viz.plot_embedding2D(X_, di_graph=G, node_colors=None)
plt.show()
# Visualize using tsne directly
X_ = TSNE(n_components=2).fit_transform(X_)
print("DEBUG::embedding vectors:")
print(X_)
pos = {}
n_nodes = X_.shape[0]
for i in range(n_nodes):
pos[i] = X_[i, :]
nx.draw_networkx(G,
pos,
node_color=kmeans.labels_,
node_size=300,
alpha=0.5,
