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
|
import argparse
import os.path as osp
import time
import torch
import torch.nn.functional as F
from torch_geometric.datasets import TUDataset
from torch_geometric.loader import DataLoader
from torch_geometric.logging import init_wandb, log
from torch_geometric.nn import MLP, GINConv, global_add_pool
parser = argparse.ArgumentParser()
parser.add_argument('--dataset', type=str, default='MUTAG')
parser.add_argument('--batch_size', type=int, default=128)
parser.add_argument('--hidden_channels', type=int, default=32)
parser.add_argument('--num_layers', type=int, default=5)
parser.add_argument('--lr', type=float, default=0.01)
parser.add_argument('--epochs', type=int, default=100)
parser.add_argument('--wandb', action='store_true', help='Track experiment')
args = parser.parse_args()
if torch.cuda.is_available():
device = torch.device('cuda')
elif hasattr(torch.backends, 'mps') and torch.backends.mps.is_available():
# MPS is currently slower than CPU due to missing int64 min/max ops
device = torch.device('cpu')
else:
device = torch.device('cpu')
init_wandb(
name=f'GIN-{args.dataset}',
batch_size=args.batch_size,
lr=args.lr,
epochs=args.epochs,
hidden_channels=args.hidden_channels,
num_layers=args.num_layers,
device=device,
)
path = osp.join(osp.dirname(osp.realpath(__file__)), '..', 'data', 'TU')
dataset = TUDataset(path, name=args.dataset).shuffle()
train_loader = DataLoader(dataset[:0.9], args.batch_size, shuffle=True)
test_loader = DataLoader(dataset[0.9:], args.batch_size)
class GIN(torch.nn.Module):
def __init__(self, in_channels, hidden_channels, out_channels, num_layers):
super().__init__()
self.convs = torch.nn.ModuleList()
for _ in range(num_layers):
mlp = MLP([in_channels, hidden_channels, hidden_channels])
self.convs.append(GINConv(nn=mlp, train_eps=False))
in_channels = hidden_channels
self.mlp = MLP([hidden_channels, hidden_channels, out_channels],
norm=None, dropout=0.5)
def forward(self, x, edge_index, batch):
for conv in self.convs:
x = conv(x, edge_index).relu()
x = global_add_pool(x, batch)
return self.mlp(x)
model = GIN(
in_channels=dataset.num_features,
hidden_channels=args.hidden_channels,
out_channels=dataset.num_classes,
num_layers=args.num_layers,
).to(device)
optimizer = torch.optim.Adam(model.parameters(), lr=args.lr)
def train():
model.train()
total_loss = 0
for data in train_loader:
data = data.to(device)
optimizer.zero_grad()
out = model(data.x, data.edge_index, data.batch)
loss = F.cross_entropy(out, data.y)
loss.backward()
optimizer.step()
total_loss += float(loss) * data.num_graphs
return total_loss / len(train_loader.dataset)
@torch.no_grad()
def test(loader):
model.eval()
total_correct = 0
for data in loader:
data = data.to(device)
out = model(data.x, data.edge_index, data.batch)
pred = out.argmax(dim=-1)
total_correct += int((pred == data.y).sum())
return total_correct / len(loader.dataset)
times = []
for epoch in range(1, args.epochs + 1):
start = time.time()
loss = train()
train_acc = test(train_loader)
test_acc = test(test_loader)
log(Epoch=epoch, Loss=loss, Train=train_acc, Test=test_acc)
times.append(time.time() - start)
print(f'Median time per epoch: {torch.tensor(times).median():.4f}s')
|
