def test_stochastic_blockmodel_graph():
torch.manual_seed(12345)
block_sizes = [2, 2, 4]
edge_probs = [
[0.25, 0.05, 0.02],
[0.05, 0.35, 0.07],
[0.02, 0.07, 0.40],
]
edge_index = stochastic_blockmodel_graph(block_sizes,
edge_probs,
directed=False)
assert edge_index.tolist() == [
[2, 3, 4, 4, 5, 5, 6, 7, 7, 7],
[3, 2, 5, 7, 4, 7, 7, 4, 5, 6],
]
edge_index = stochastic_blockmodel_graph(block_sizes,
edge_probs,
directed=True)
assert edge_index.tolist() == [
[0, 1, 3, 5, 6, 6, 7, 7],
[3, 3, 2, 4, 4, 7, 5, 6],
]
def process(self):
from sklearn.datasets import make_classification
edge_index = stochastic_blockmodel_graph(
self.block_sizes, self.edge_probs, directed=not self.is_undirected)
num_samples = int(self.block_sizes.sum())
num_classes = self.block_sizes.size(0)
x = None
if self.num_channels is not None:
x, _ = make_classification(
n_samples=num_samples,
n_features=self.num_channels,
n_classes=num_classes,
weights=self.block_sizes / num_samples,
**self.kwargs,
)
x = torch.from_numpy(x).to(torch.float)
y = torch.arange(num_classes).repeat_interleave(self.block_sizes)
data = Data(x=x, edge_index=edge_index, y=y)
if self.pre_transform is not None:
data = self.pre_transform(data)
torch.save(self.collate([data]), self.processed_paths[0])
for y in range(len(block_sizes)):
cur_prob = 0.0
if x == y:
cur_prob = INTER_PROB[x]
elif x >= NUM_BLOCKS or y >= NUM_BLOCKS:
cur_prob = NOISE_PROB
else:
if USE_NOISE_NODES:
cur_prob = INTRA_PROB
else:
cur_prob = INTRA_PROB + sbm_noise
cur_probs.append(float(cur_prob))
edge_probs.append(cur_probs)
sbm_ei = stochastic_blockmodel_graph(block_sizes,
edge_probs,
directed=False)
sbm_x = []
for x in range((COMMUNITY_SIZE * NUM_BLOCKS) + len(block_sizes) -
NUM_BLOCKS):
sbm_x.append([1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0])
sbm_x = torch.tensor(sbm_x)
sbm_y = []
for ind, x in enumerate(block_sizes):
for y in range(x):
sbm_y.append(int(ind))
sbm_y = torch.tensor(sbm_y)
print(sbm_ei)
print(sbm_x)
print(sbm_y)
real_data.append(