def get_tg_dataset(args, dataset_name, use_cache=True, remove_feature=False):
# "Cora", "CiteSeer" and "PubMed"
if dataset_name in ['Cora', 'CiteSeer', 'PubMed']:
dataset = tg.datasets.Planetoid(root='datasets/' + dataset_name,
name=dataset_name)
else:
try:
dataset = load_tg_dataset(dataset_name)
except:
raise NotImplementedError
# precompute shortest path
if not os.path.isdir('datasets'):
os.mkdir('datasets')
if not os.path.isdir('datasets/cache'):
os.mkdir('datasets/cache')
f1_name = 'datasets/cache/' + dataset_name + str(
args.approximate) + '_dists.dat'
f2_name = 'datasets/cache/' + dataset_name + str(
args.approximate) + '_dists_removed.dat'
f3_name = 'datasets/cache/' + dataset_name + str(
args.approximate) + '_links_train.dat'
f4_name = 'datasets/cache/' + dataset_name + str(
args.approximate) + '_links_val.dat'
f5_name = 'datasets/cache/' + dataset_name + str(
args.approximate) + '_links_test.dat'
if use_cache and ((os.path.isfile(f2_name) and args.task == 'link') or
(os.path.isfile(f1_name) and args.task != 'link')):
with open(f3_name, 'rb') as f3, \
open(f4_name, 'rb') as f4, \
open(f5_name, 'rb') as f5:
links_train_list = pickle.load(f3)
links_val_list = pickle.load(f4)
links_test_list = pickle.load(f5)
if args.task == 'link':
with open(f2_name, 'rb') as f2:
dists_removed_list = pickle.load(f2)
else:
with open(f1_name, 'rb') as f1:
dists_list = pickle.load(f1)
print('Cache loaded!')
data_list = []
for i, data in enumerate(dataset):
if args.task == 'link':
data.mask_link_positive = deduplicate_edges(
data.edge_index.numpy())
data.mask_link_positive_train = links_train_list[i]
data.mask_link_positive_val = links_val_list[i]
data.mask_link_positive_test = links_test_list[i]
get_link_mask(data, resplit=False)
if args.task == 'link':
data.dists = torch.from_numpy(dists_removed_list[i]).float()
data.edge_index = torch.from_numpy(
duplicate_edges(data.mask_link_positive_train)).long()
else:
data.dists = torch.from_numpy(dists_list[i]).float()
if remove_feature:
data.x = torch.ones((data.x.shape[0], 1))
data_list.append(data)
else:
data_list = []
dists_list = []
dists_removed_list = []
links_train_list = []
links_val_list = []
links_test_list = []
for i, data in enumerate(dataset):
if 'link' in args.task:
print(f"args.task = {args.task}")
get_link_mask(
data,
args.remove_link_ratio,
resplit=True,
infer_link_positive=True if args.task == 'link' else False)
links_train_list.append(data.mask_link_positive_train)
links_val_list.append(data.mask_link_positive_val)
links_test_list.append(data.mask_link_positive_test)
if args.task == 'link':
dists_removed = precompute_dist_data(
data.mask_link_positive_train,
data.num_nodes,
approximate=args.approximate)
dists_removed_list.append(dists_removed)
data.dists = torch.from_numpy(dists_removed).float()
data.edge_index = torch.from_numpy(
duplicate_edges(data.mask_link_positive_train)).long()
else:
dists = precompute_dist_data(data.edge_index.numpy(),
data.num_nodes,
approximate=args.approximate)
dists_list.append(dists)
data.dists = torch.from_numpy(dists).float()
if remove_feature:
data.x = torch.ones((data.x.shape[0], 1))
data_list.append(data)
with open(f1_name, 'wb') as f1, \
open(f2_name, 'wb') as f2, \
open(f3_name, 'wb') as f3, \
open(f4_name, 'wb') as f4, \
open(f5_name, 'wb') as f5:
if args.task == 'link':
pickle.dump(dists_removed_list, f2)
else:
pickle.dump(dists_list, f1)
pickle.dump(links_train_list, f3)
pickle.dump(links_val_list, f4)
pickle.dump(links_test_list, f5)
print('Cache saved!')
return data_list
def get_tg_dataset(args, dataset_name, use_cache=True, remove_feature=False, hash_overwrite=False, hash_concat=False):
# "Cora", "CiteSeer" and "PubMed"
if dataset_name in ['Cora', 'CiteSeer', 'PubMed']:
dataset = tg.datasets.Planetoid(root='datasets/' + dataset_name, name=dataset_name)
else:
try:
dataset = load_tg_dataset(dataset_name)
except:
raise NotImplementedError
# precompute shortest path
if not os.path.isdir('datasets/cache'):
os.mkdir('datasets/cache')
f1_name = 'datasets/cache/' + dataset_name + str(args.approximate) + '_dists.dat'
f2_name = 'datasets/cache/' + dataset_name + str(args.approximate)+ '_dists_removed.dat'
f3_name = 'datasets/cache/' + dataset_name + str(args.approximate)+ '_links_train.dat'
f4_name = 'datasets/cache/' + dataset_name + str(args.approximate)+ '_links_val.dat'
f5_name = 'datasets/cache/' + dataset_name + str(args.approximate)+ '_links_test.dat'
# cache for dists_all
f6_name = 'datasets/cache/' + dataset_name + str(args.approximate)+ '_dists_all.dat'
if use_cache and ((os.path.isfile(f2_name) and args.task=='link') or (os.path.isfile(f1_name) and
args.task != 'link')):
with open(f3_name, 'rb') as f3, \
open(f4_name, 'rb') as f4, \
open(f5_name, 'rb') as f5, \
open(f6_name, 'rb') as f6:
links_train_list = pickle.load(f3)
links_val_list = pickle.load(f4)
links_test_list = pickle.load(f5)
# load a list of dists_all (each is for one connected components)
dists_all_list = pickle.load(f6)
if args.task=='link':
with open(f2_name, 'rb') as f2:
dists_removed_list = pickle.load(f2)
else:
with open(f1_name, 'rb') as f1:
dists_list = pickle.load(f1)
print('Cache loaded!')
data_list = []
start_node = 0
for i, data in enumerate(dataset):
if args.task == 'link':
data.mask_link_positive = deduplicate_edges(data.edge_index.numpy())
data.mask_link_positive_train = links_train_list[i]
data.mask_link_positive_val = links_val_list[i]
data.mask_link_positive_test = links_test_list[i]
get_link_mask(data, resplit=False)
if args.task == 'link':
data.dists = torch.from_numpy(dists_removed_list[i]).float()
data.edge_index = torch.from_numpy(duplicate_edges(data.mask_link_positive_train)).long()
else:
data.dists = torch.from_numpy(dists_list[i]).float()
if remove_feature:
data.x = torch.ones((data.x.shape[0],1))
if hash_overwrite:
x = np.zeros(data.x.shape)
for m in range(data.x.shape[0]):
x[m] = int_to_hash_vector(start_node + m, data.x.shape[1])
data.x = torch.from_numpy(x).toFloat()
start_node += data.x.shape[0]
if hash_concat:
x = np.zeros((data.x.shape[0], data.x.shape[1] * 2))
for m in range(data.x.shape[0]):
x[m] = np.concatenate((data.x[m], int_to_hash_vector(start_node + m, data.x.shape[1])))
data.x = torch.from_numpy(x).toFloat()
start_node += data.x.shape[0]
# assign dists_all to each data (connected components)
data.dists_all = dists_all_list[i]
# generate graph dist ranks
data.dists_ranks = gen_graph_dist_rank_data(data.dists_all)
data_list.append(data)
else:
dists_all_list = [] # dists_all stores dists for all nodes (regardless whether it's in train, val or test)
data_list = []
dists_list = []
dists_removed_list = []
links_train_list = []
links_val_list = []
links_test_list = []
start_node = 0
for i, data in enumerate(dataset):
if 'link' in args.task:
get_link_mask(data, args.remove_link_ratio, resplit=True,
infer_link_positive=True if args.task == 'link' else False)
links_train_list.append(data.mask_link_positive_train)
links_val_list.append(data.mask_link_positive_val)
links_test_list.append(data.mask_link_positive_test)
if args.task=='link':
dists_removed = precompute_dist_data(data.mask_link_positive_train, data.num_nodes,
approximate=args.approximate)
dists_removed_list.append(dists_removed)
data.dists = torch.from_numpy(dists_removed).float()
data.edge_index = torch.from_numpy(duplicate_edges(data.mask_link_positive_train)).long()
else:
dists = precompute_dist_data(data.edge_index.numpy(), data.num_nodes, approximate=args.approximate)
dists_list.append(dists)
data.dists = torch.from_numpy(dists).float()
# calculate dists for all nodes in the connected component, no need to worry about if the task is 'link'
dists_all = precompute_dist_data(data.edge_index.numpy(), data.num_nodes, approximate=args.approximate)
dists_all_list.append(dists_all)
data.dists_all = dists_all
if remove_feature:
data.x = torch.ones((data.x.shape[0],1))
if hash_overwrite:
x = np.zeros(data.x.shape)
for m in range(data.x.shape[0]):
x[m] = int_to_hash_vector(start_node + m, data.x.shape[1])
data.x = torch.from_numpy(x).float()
start_node += data.x.shape[0]
if hash_concat:
x = np.zeros((data.x.shape[0], data.x.shape[1] * 2))
for m in range(data.x.shape[0]):
x[m] = np.concatenate((data.x[m], int_to_hash_vector(start_node + m, data.x.shape[1])))
data.x = torch.from_numpy(x).float()
start_node += data.x.shape[0]
# generate graph dist ranks
data.dists_ranks = gen_graph_dist_rank_data(data.dists_all)
data_list.append(data)
with open(f1_name, 'wb') as f1, \
open(f2_name, 'wb') as f2, \
open(f3_name, 'wb') as f3, \
open(f4_name, 'wb') as f4, \
open(f5_name, 'wb') as f5, \
open(f6_name, 'wb') as f6:
if args.task == 'link':
pickle.dump(dists_removed_list, f2)
else:
pickle.dump(dists_list, f1)
pickle.dump(links_train_list, f3)
pickle.dump(links_val_list, f4)
pickle.dump(links_test_list, f5)
pickle.dump(dists_all_list, f6)
print('Cache saved!')
return data_list