-
Notifications
You must be signed in to change notification settings - Fork 7
/
Copy pathtrain_linkpred_ogb.py
198 lines (156 loc) · 8.87 KB
/
train_linkpred_ogb.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
import argparse
from copy import copy
from tqdm.auto import tqdm
import torch
import torch_geometric.transforms as T
from torch_geometric.utils import to_undirected
from ogb.linkproppred import Evaluator, PygLinkPropPredDataset
# custom modules
from maskgae.utils import set_seed, tab_printer
from maskgae.model import MaskGAE, DegreeDecoder, EdgeDecoder, GNNEncoder, DotEdgeDecoder
from maskgae.mask import MaskEdge, MaskPath
def train_linkpred(model, splits, args, device="cpu"):
optimizer = torch.optim.Adam(model.parameters(),
lr=args.lr,
weight_decay=args.weight_decay)
best_valid = 0
batch_size = args.batch_size
train_data = splits['train'].to(device)
valid_data = splits['valid'].to(device)
test_data = splits['test'].to(device)
monitor = 'Hits@50'
evaluator = Evaluator(name=args.dataset)
for epoch in tqdm(range(1, 1 + args.epochs)):
loss = model.train_step(train_data, optimizer,
alpha=args.alpha,
batch_size=args.batch_size)
if epoch % args.eval_period == 0:
valid_results = model.test_step_ogb(valid_data, evaluator,
valid_data.pos_edge_label_index,
valid_data.neg_edge_label_index,
batch_size=batch_size)
test_results = model.test_step_ogb(test_data, evaluator,
valid_data.pos_edge_label_index,
valid_data.neg_edge_label_index,
batch_size=batch_size)
if valid_results[monitor] > best_valid:
best_valid = valid_results[monitor]
torch.save(model.state_dict(), args.save_path)
print()
print(f"Epoch {epoch} - Hits@20: {test_results['Hits@20']:.2%}",
f"Hits@50: {test_results['Hits@50']:.2%}",
f"Hits@100: {test_results['Hits@100']:.2%}")
model.load_state_dict(torch.load(args.save_path))
results = model.test_step_ogb(test_data, evaluator,
valid_data.pos_edge_label_index,
valid_data.neg_edge_label_index,
batch_size=batch_size)
return results
parser = argparse.ArgumentParser()
parser.add_argument("--dataset", nargs="?", default="ogbl-collab", help="Datasets. (default: ogbl-collab)")
parser.add_argument("--mask", nargs="?", default="Path", help="Masking stractegy, `Path`, `Edge` or `None` (default: Path)")
parser.add_argument('--seed', type=int, default=2022, help='Random seed for model and dataset. (default: 2022)')
parser.add_argument('--bn', action='store_true', help='Whether to use batch normalization for GNN encoder. (default: False)')
parser.add_argument("--layer", nargs="?", default="sage", help="GNN layer, (default: sage)")
parser.add_argument("--encoder_activation", nargs="?", default="elu", help="Activation function for GNN encoder, (default: elu)")
parser.add_argument('--encoder_channels', type=int, default=256, help='Channels of GNN encoder. (default: 256)')
parser.add_argument('--hidden_channels', type=int, default=256, help='Channels of hidden representation. (default: 256)')
parser.add_argument('--decoder_channels', type=int, default=64, help='Channels of decoder. (default: 64)')
parser.add_argument('--encoder_layers', type=int, default=1, help='Number of layers of encoder. (default: 1)')
parser.add_argument('--decoder_layers', type=int, default=2, help='Number of layers for decoders. (default: 2)')
parser.add_argument('--encoder_dropout', type=float, default=0.3, help='Dropout probability of encoder. (default: 0.3)')
parser.add_argument('--decoder_dropout', type=float, default=0.3, help='Dropout probability of decoder. (default: 0.3)')
parser.add_argument('--alpha', type=float, default=0.003, help='loss weight for degree prediction. (default: 2e-3)')
parser.add_argument('--lr', type=float, default=0.001, help='Learning rate for training. (default: 0.001)')
parser.add_argument('--weight_decay', type=float, default=0., help='weight_decay for training. (default: 0.)')
parser.add_argument('--grad_norm', type=float, default=1.0, help='grad_norm for training. (default: 1.0.)')
parser.add_argument('--batch_size', type=int, default=2**16, help='Number of batch size. (default: 2**16)')
parser.add_argument("--start", nargs="?", default="edge", help="Which Type to sample starting nodes for random walks, (default: edge)")
parser.add_argument('--p', type=float, default=0.7, help='Mask ratio or sample ratio for MaskEdge/MaskPath')
parser.add_argument('--epochs', type=int, default=200, help='Number of training epochs. (default: 200)')
parser.add_argument('--runs', type=int, default=10, help='Number of runs. (default: 10)')
parser.add_argument('--eval_period', type=int, default=10, help='(default: 10)')
parser.add_argument("--save_path", nargs="?", default="MaskGAE-OGB.pt", help="save path for model. (default: MaskGAE-OGB.pt)")
parser.add_argument("--device", type=int, default=0)
try:
args = parser.parse_args()
print(tab_printer(args))
except:
parser.print_help()
exit(0)
set_seed(args.seed)
if args.device < 0:
device = "cpu"
else:
device = f"cuda:{args.device}" if torch.cuda.is_available() else "cpu"
transform = T.Compose([
# T.ToUndirected(),
T.ToDevice(device),
])
# (!IMPORTANT) Specify the path to your dataset directory ##############
# root = '~/public_data/pyg_data' # my root directory
root = 'data/'
########################################################################
print('Loading Data...')
if args.dataset in {'ogbl-collab'}:
dataset = PygLinkPropPredDataset(name=args.dataset, root=root)
data = transform(dataset[0])
del data.edge_weight, data.edge_year
else:
raise ValueError(args.dataset)
split_edge = dataset.get_edge_split()
args.year = 2010
if args.year > 0:
year_mask = split_edge['train']['year'] >= args.year
split_edge['train']['edge'] = split_edge['train']['edge'][year_mask]
data.edge_index = to_undirected(split_edge['train']['edge'].t())
print(f"{1 - year_mask.float().mean():.2%} of edges are dropped accordding to edge year {args.year}.")
train_data, val_data, test_data = copy(data), copy(data), copy(data)
args.val_as_input = True
if args.val_as_input:
full_edge_index = torch.cat([split_edge['train']['edge'], split_edge['valid']['edge']], dim=0).t()
full_edge_index = to_undirected(full_edge_index)
train_data.edge_index = full_edge_index
val_data.edge_index = full_edge_index
test_data.edge_index = full_edge_index
train_data.pos_edge_label_index = torch.cat([split_edge['train']['edge'], split_edge['valid']['edge']], dim=0).t()
else:
train_data.pos_edge_label_index = split_edge['train']['edge'].t()
val_data.pos_edge_label_index = split_edge['valid']['edge'].t()
val_data.neg_edge_label_index = split_edge['valid']['edge_neg'].t()
test_data.pos_edge_label_index = split_edge['test']['edge'].t()
test_data.neg_edge_label_index = split_edge['test']['edge_neg'].t()
splits = dict(train=train_data, valid=val_data, test=test_data)
if args.mask == 'Path':
mask = MaskPath(p=args.p, num_nodes=data.num_nodes,
start=args.start,
walk_length=args.encoder_layers+1)
elif args.mask == 'Edge':
mask = MaskEdge(p=args.p)
else:
mask = None # vanilla GAE
encoder = GNNEncoder(data.num_features, args.encoder_channels, args.hidden_channels,
num_layers=args.encoder_layers, dropout=args.encoder_dropout,
bn=args.bn, layer=args.layer, activation=args.encoder_activation,
use_node_feats=False, node_emb=256, num_nodes=data.num_nodes)
edge_decoder = DotEdgeDecoder()
degree_decoder = DegreeDecoder(args.hidden_channels, args.decoder_channels,
num_layers=args.decoder_layers, dropout=args.decoder_dropout)
model = MaskGAE(encoder, edge_decoder, degree_decoder, mask,
random_negative_sampling=True, loss='auc').to(device)
hit_20 = []
hit_50 = []
hit_100 = []
for run in range(1, args.runs+1):
hit = train_linkpred(model, splits, args, device=device)
hit_20.append(hit['Hits@20'])
hit_50.append(hit['Hits@50'])
hit_100.append(hit['Hits@100'])
print(f"Runs {run} - Hits@20: {hit['Hits@20']:.2%}",
f"Hits@50: {hit['Hits@50']:.2%}",
f"Hits@100: {hit['Hits@100']:.2%}")
print(f'Link Prediction Results ({args.runs} runs):\n'
f'Hits@20: {np.mean(hit_20):.2%} ± {np.std(hit_20):.2%}',
f'Hits@50: {np.mean(hit_50):.2%} ± {np.std(hit_50):.2%}',
f'Hits@100: {np.mean(hit_100):.2%} ± {np.std(hit_100):.2%}',
)