def convert_sparse_train_input1(adj, features):
supports = preprocess_adj(adj)
m = torch.from_numpy(supports[0]).long()
n = torch.from_numpy(supports[1])
support = torch.sparse.FloatTensor(m.t(), n, supports[2]).float()
features = [torch.tensor(idxs, dtype=torch.long).to(device) if torch.cuda.is_available() else \
torch.tensor(idxs, dtype=torch.long) for idxs in features]
# i = torch.from_numpy(features[0]).long()
# v = torch.from_numpy(features[1])
# feature = torch.sparse.FloatTensor(i.t(), v, features[2])
if torch.cuda.is_available():
m = m.to(device)
n = n.to(device)
# support = torch.sparse.FloatTensor(m.t(), n, supports[2]).float().to(device)
support = torch.sparse.FloatTensor(m.t(), n, supports[2]).float().to(device)
# i = i.to(device)
# v = v.to(device)
# feature = torch.sparse.FloatTensor(i.t(), v, features[2]).to(device)
# print("**", features[0].dtype)
# print("**", support.dtype)
return features, support
seed = 123
np.random.seed(seed)
torch.random.manual_seed(seed)
# load data
adj, features, y_train, y_val, y_test, train_mask, val_mask, test_mask = load_data(
args.dataset)
print('adj:', adj.shape)
print('features:', features.shape)
print('y:', y_train.shape, y_val.shape, y_test.shape)
print('mask:', train_mask.shape, val_mask.shape, test_mask.shape)
# D^-1@X
features = preprocess_features(features) # [49216, 2], [49216], [2708, 1433]
supports = preprocess_adj(adj)
# D^-1@X
features = preprocess_features(features) # [49216, 2], [49216], [2708, 1433]
supports = preprocess_adj(adj)
# 数据放到cuda
device = torch.device('cuda')
train_label = torch.from_numpy(y_train).long().to(device)
num_classes = train_label.shape[1]
train_label = train_label.argmax(dim=1)
train_mask = torch.from_numpy(train_mask.astype(np.int)).to(device)
val_label = torch.from_numpy(y_val).long().to(device)
val_label = val_label.argmax(dim=1)
val_mask = torch.from_numpy(val_mask.astype(np.int)).to(device)