def load_tugraphs(graph='mutag', labels_only = False):
# load synthetic sbm graphs
if graph[:3] == 'syn':
scale = int(graph[3:])
n = 750
p, q = 0.5, 0.1
p_, q_ = 0.4, 0.2
gs1 = sbms(n=n, n1=100 * scale, n2=50 * scale, p=p, q=q)
gs2 = sbms(n=n, n1=75 * scale, n2=75 * scale, p=p, q=q)
graphs = gs2 + gs1
labels = [1] * n + [2] * n
return graphs, labels
graph = name_conversion(graph)
path = osp.join(osp.dirname(osp.realpath(__file__)), '..', 'data', graph)
dataset = TUDataset(path, name=graph)
graphs, labels = torch_geometric_2nx(dataset, labels_only)
return graphs, labels
""" sbm graph classification """
from Esme.dgms.fil import nodefeat
from Esme.graph.function import fil_strategy
from Esme.graph.generativemodel import sbms
from networkx.linalg.algebraicconnectivity import fiedler_vector
if __name__ == '__main__':
n = 1
p, q = 0.5, 0.1
gs = sbms(n=n, n1=100, n2=50, p=p, q=q)
for i in range(len(gs)):
g = gs[i]
# lapfeat = nodefeat(g, 'fiedler', norm=True)
nodefeat = fiedler_vector(g, normalized=False) # np.ndarray
nodefeat = nodefeat.reshape(len(g), 1)
gs[i] = fil_strategy(g, nodefeat, method='node', viz_flag=False)
print('Finish computing lapfeat')
parser.add_argument("--q",
default=0.1,
type=float,
help='The probablity btwn community')
if __name__ == '__main__':
# sys.argv = ['graph/2sbm_gc.py']
args = parser.parse_args()
print(args)
n = 100
p, q = 0.5, args.q
p_, q_ = 0.4, 0.2
fil_method = args.fil_method
zigzag = True if fil_method == 'combined' else False
edge_kwargs = {'h': 0.3, 'edgefunc': 'edge_prob'}
gs1 = sbms(n=n, n1=100, n2=50, p=p, q=q)
gs2 = sbms(n=n, n1=75, n2=75, p=p, q=q)
# gs3 = sbms(n=n, n1=75, n2=75, p=p_, q=q_)
# gs3 = sbms(n=n, n1=50, n2=50, p=p, q=q)
gs = gs2 + gs1
labels = [1] * n + [2] * n
plt.title('p: %s, q: %s' % (p, q))
for i in range(len(gs)):
g = gs[i]
lp = LaplacianEigenmaps(d=1)
lp.learn_embedding(g, weight='weight')
lapfeat = lp.get_embedding() # lapfeat is an array
# viz = True if i%100==0 else False
# plt.subplot(2, 1, 1+i//100)