-
Notifications
You must be signed in to change notification settings - Fork 1
/
rgcn_classifier_heterographs.py
143 lines (117 loc) · 5.56 KB
/
rgcn_classifier_heterographs.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
import dgl
def collate(samples):
# The input `samples` is a list of pairs
# (graph, label).
graphs, labels = map(list, zip(*samples))
batched_graph = dgl.batch(graphs)
return batched_graph, torch.tensor(labels)
import dgl.function as fn
import dgl.nn.pytorch as dglnn
import torch
import torch.nn as nn
import torch.nn.functional as F
# Sends a message of node feature h.
msg = fn.copy_src(src='features', out='m')
def reduce(nodes):
"""Take an average over all neighbor node features hu and use it to
overwrite the original node feature."""
accum = torch.mean(nodes.mailbox['m'], 1)
return {'features': accum}
class RGCN(nn.Module):
def __init__(self, in_feats, hid_feats, out_feats, rel_names):
super().__init__()
self.conv1 = dglnn.HeteroGraphConv({
rel: dglnn.GraphConv(in_feats, hid_feats)
for rel in rel_names}, aggregate='sum')
self.conv2 = dglnn.HeteroGraphConv({
rel: dglnn.GraphConv(hid_feats, out_feats)
for rel in rel_names}, aggregate='sum')
def forward(self, graph, inputs):
# inputs is features of nodes
h = self.conv1(graph, inputs)
h = {k: F.relu(v) for k, v in h.items()}
h = self.conv2(graph, h)
return h
class HeteroClassifier(nn.Module):
def __init__(self, in_dim, hidden_dim, n_classes, rel_names):
super().__init__()
self.rgcn = RGCN(in_dim, hidden_dim, hidden_dim, rel_names)
self.classify = nn.Linear(hidden_dim, n_classes)
def forward(self, g):
h = g.ndata['features']
h = self.rgcn(g, h)
with g.local_scope():
g.ndata['features'] = h
# Calculate graph representation by average readout.
hg = 0
for ntype in g.ntypes:
hg = hg + dgl.mean_nodes(g, 'features', ntype=ntype)
return self.classify(hg)
import torch.optim as optim
from torch.utils.data import DataLoader
def main(bug_type, use_deepbugs_embeddings, dataset_size):
print('----RGCN Training on hetero graphs in bug type {} with {}----'.format(bug_type, 'deepbugs embeddings' if use_deepbugs_embeddings else 'random embeddings'))
# Create training and test sets.
if dataset_size == 'mini':
from heterogenous_mini_dataset import MiniCorrectAndBuggyDataset
trainset = MiniCorrectAndBuggyDataset(use_deepbugs_embeddings=use_deepbugs_embeddings, is_training=True, bug_type=bug_type)
testset = MiniCorrectAndBuggyDataset(use_deepbugs_embeddings=use_deepbugs_embeddings, is_training=False, bug_type=bug_type)
elif dataset_size == 'full':
from heterogenous_full_dataset import FullCorrectAndBuggyDataset
trainset = FullCorrectAndBuggyDataset(use_deepbugs_embeddings=use_deepbugs_embeddings, is_training=True, bug_type=bug_type)
testset = FullCorrectAndBuggyDataset(use_deepbugs_embeddings=use_deepbugs_embeddings, is_training=False, bug_type=bug_type)
# Use PyTorch's DataLoader and the collate function
# defined before.
data_loader = DataLoader(trainset, batch_size=100, shuffle=True,
collate_fn=collate)
def evaluate():
## Evaluate model
model.eval()
# Convert a list of tuples to two lists
test_X, test_Y = map(list, zip(*testset))
test_bg = dgl.batch(test_X)
test_Y = torch.tensor(test_Y).float().view(-1, 1)
prediction = model(test_bg)
probs_Y = torch.softmax(prediction, 1)
sampled_Y = torch.multinomial(probs_Y, 1)
argmax_Y = torch.max(probs_Y, 1)[1].view(-1, 1)
print('Accuracy of sampled predictions on the test set: {:.4f}%'.format(
(test_Y == sampled_Y.float()).sum().item() / len(test_Y) * 100))
print('Accuracy of argmax predictions on the test set: {:4f}%'.format(
(test_Y == argmax_Y.float()).sum().item() / len(test_Y) * 100))
# Create model
model = HeteroClassifier(200, 16, trainset.num_classes, \
['follows', 'has_value_of', 'has_value_type', 'precedes', 'is_left_param_of', 'is_right_param_of', 'is_type_of'] if bug_type == 'swapped_args' else ['parent', 'parent', 'precedes', 'parent', 'followedBy', 'follows', 'typeOf'])
loss_func = nn.CrossEntropyLoss()
optimizer = optim.Adam(model.parameters(), lr=0.005)
model.train()
epoch_losses = []
for epoch in range(30):
epoch_loss = 0
for iter, (bg, label) in enumerate(data_loader):
prediction = model(bg)
loss = loss_func(prediction, label)
optimizer.zero_grad()
loss.backward()
optimizer.step()
epoch_loss += loss.detach().item()
epoch_loss /= (iter + 1)
print('Epoch {}, loss {:.4f}'.format(epoch, epoch_loss))
epoch_losses.append(epoch_loss)
if epoch % 5 == 0:
evaluate()
evaluate()
import argparse
parser = argparse.ArgumentParser()
parser.add_argument(
'--bug_type', help='Type of bug to train', choices=['swapped_args', 'binOps', 'incorrect_binary_operator', 'incorrect_binary_operand'], required=False)
parser.add_argument(
'--use_deepbugs_embeddings', help='Random or deepbugs embeddings', required=False)
parser.add_argument(
'--dataset_size', help='Mini or Full dataset', choices=['mini', 'full'], required=False)
if __name__=='__main__':
args = parser.parse_args()
bug_type = args.bug_type or 'all'
use_deepbugs_embeddings = True if args.use_deepbugs_embeddings in ['True', 'true'] else False
dataset_size = args.dataset_size or 'mini'
main(bug_type, use_deepbugs_embeddings, dataset_size)