-
Notifications
You must be signed in to change notification settings - Fork 1
/
train.py
164 lines (128 loc) · 5.19 KB
/
train.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
# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import numpy as np
import paddle
from net import L0_SIGN
from paddle.optimizer import Adam
import pgl
from pgl.utils.logger import log
from pgl.utils.data import Dataloader
from dataset import SIGNDataset, random_split, collate_fn
def train(args, data_info, data_nums):
paddle.device.set_device(args.device)
train_loader = data_info[0]
val_loader = data_info[1]
test_loader = data_info[2]
num_feature = data_info[3]
model = L0_SIGN(args, num_feature)
# print(model.sublayers())
# print('----'*20)
# for item in model.named_sublayers():
# print(item)
# print('===='*20)
# print(model.parameters())
# optimizer = torch.optim.Adagrad(
# filter(lambda p: p.requires_grad, model.parameters()),
# args.lr,
# lr_decay=1e-5,
# #weight_decay=1e-5
# )
optimizer = paddle.optimizer.Adagrad(learning_rate=args.lr,
parameters=model.parameters(),epsilon=1e-05,weight_decay=1e-05)
# optimizer =paddle.optimizer.Adam(learning_rate=args.lr,
# parameters=model.parameters())
# crit = paddle.nn.BCELoss()
crit = paddle.nn.MSELoss()
# crit = paddle.nn.CrossEntropyLoss()
# print([i.size() for i in filter(lambda p: not p.stop_gradient, model.parameters())])
log.info('start training...')
for step in range(args.n_epoch):
# training
loss_all = 0
edge_all = 0
model.train()
n = 0
for data in train_loader:
n += 1
g, label = data
g = pgl.Graph.batch(g).tensor()
label = paddle.to_tensor(label,dtype='float32')
# print(g.graph_node_id)
#return_dict = model(*get_feed_dict(args, model, train_data, ripple_set, start, start + args.batch_size))
output, l0_penaty, l2_penaty, num_edges = model(g)
# print('===='*8)
# print('output:',output)
# print('label:',label)
# label = label[:,1].reshape((-1,1))
# label = paddle.to_tensor(label,dtype='int64')
baseloss = crit(output, label)
l0_loss = l0_penaty * args.l0_weight
l2_loss = l2_penaty * args.l2_weight
loss = baseloss + l0_loss + l2_loss
loss_all += g.num_graph * loss.detach().item()
loss.backward()
# for k in model.parameters():
# if n<2:
# print('=='*20)
# print(k.grad)
optimizer.step()
optimizer.clear_grad()
cur_loss = loss_all / data_nums[0]
# evaluation
# train_auc = 0
train_auc, train_acc, train_edges = evaluate(model, train_loader)
# train_auc, train_acc, train_edges = 0, 0, 0
val_auc,val_acc, _ = evaluate(model, val_loader)
test_auc, test_acc, test_edges = evaluate(model, test_loader)
# print(cur_loss)
log.info('Epoch: {:03d}, Loss: {:.4f}, Train Auc: {:.4f},Train Acc: {:.4f},Train edges: {:07d}, Val Auc: {:.4f}, Acc: {:.4f}, Test Auc: {:.4f}, Acc: {:.4f}, Train edges: {:07d}'.
format(step, cur_loss.numpy()[0].astype('float32'), train_auc,train_acc,train_edges, val_auc, val_acc, test_auc, test_acc, test_edges))
# .numpy()[0].astype('float32')
def evaluate(model, loader):
model.eval()
predictions = []
labels = []
edges_all = 0
m_auc = paddle.metric.Auc()
m_acc = paddle.metric.Accuracy()
m_auc.reset()
m_acc.reset()
with paddle.no_grad():
for data in loader:
g, label = data
g = pgl.Graph.batch(g).tensor()
label = paddle.to_tensor(label,dtype='float32')
pred, _, _, num_edges = model(g)
# print(g.num_graph)
# print(num_edges)
# print('==='*15)
pred = pred.detach().cpu().numpy()
edges_all += num_edges
predictions.append(pred)
labels.append(label)
predictions = np.vstack(predictions)
labels = np.vstack(labels)
labels = labels[:,1].reshape((-1,1))
m_auc.update(preds=predictions, labels=labels)
auc = m_auc.accumulate()
# print('predictions:',predictions)
# print('labels:',labels)
predictions = paddle.to_tensor(predictions,dtype='float32')
labels = paddle.to_tensor(labels,dtype='int64')
correct = m_acc.compute(predictions, labels)
m_acc.update(correct)
acc = m_acc.accumulate()
# auc = roc_auc_score(labels, predictions)
# acc = accuracy_score(np.argmax(labels, 1), np.argmax(predictions, 1))
return auc, acc, edges_all