default=2,
help='number of layers for Hypergraph (default: 2)')
parser.add_argument('--num_blocks', type=int, default=2, help='num_blocks')
parser.add_argument('--num_heads', type=int, default=1, help='num_heads')
parser.add_argument(
'--pos_fixed',
type=int,
default=0,
help='trainable positional embedding usually has better performance')
args = parser.parse_args()
tf.set_random_seed(args.seed)
args.neg_size = 1
train_set, val_set, train_val_set, test_set, data_time, neg_test, num_items, num_users = Dataset.data_partition_neg(
args)
print(data_time[-1])
print(data_time[-2])
subgraphs_mapping_i, subgraphs_G, subgraphs_mapping_u = hgut.subgraph_con(
train_set, data_time[0], data_time[-2])
subgraphs_mapping_i, reversed_subgraphs_mapping_i, sorted_time, subgraphs_sequence_i, reversed_subgraphs_mapping_last_i = hgut.subgraph_key_building(
subgraphs_mapping_i, num_items)
subgraphs_mapping_u, reversed_subgraphs_mapping_u, sorted_time_u, subgraphs_sequence_u, reversed_subgraphs_mapping_last_u = hgut.subgraph_key_building(
subgraphs_mapping_u, num_users)
assert sorted_time == sorted_time_u
