-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathzdsw.py
266 lines (244 loc) · 10.2 KB
/
zdsw.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
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
import torch
from torch_geometric.data import Data, DataLoader, InMemoryDataset
import argparse
import torch.nn.functional as F
from torch_geometric.nn import GCNConv, ChebConv, NNConv, GATConv
from torch_geometric.nn import SAGPooling
from torch_geometric.nn import global_mean_pool as gap, global_max_pool as gmp
from torch import nn
from zdsw_Dataset import zdswDataset
import os
import matplotlib.pyplot as plt
from smoothing import LabelSmoothingCrossEntropy, smooth_one_hot, LabelSmoothingLoss, ECELoss
from sklearn.metrics import f1_score
import scipy.io as sio
import time
from torch_scatter import scatter_mean, scatter_max
from torchsummary import summary
torch.cuda.set_device(2)
os.environ['CUDA_VISIBLE_DEVICES'] = '2'
parser = argparse.ArgumentParser()
parser.add_argument('--batch_size', type=int, default=64,
help='batch size')
parser.add_argument('--lr', type=float, default=0.001,
help='learning rate')
parser.add_argument('--momentum', type=float, default=0.9,
help='momentum')
parser.add_argument('--weight_decay', type=float, default=0.0005,
help='weight decay')
parser.add_argument('--nhid', type=int, default=128,
help='hidden size')
parser.add_argument('--pooling_ratio', type=float, default=0.5,
help='pooling ratio')
parser.add_argument('--dropout_ratio', type=float, default=0.4,
help='dropout ratio')
parser.add_argument('--epochs', type=int, default=500,
help='maximum number of epochs')
parser.add_argument('--patience', type=int, default=50,
help='patience for early stop')
parser.add_argument('--pooling_layer_type', type=str, default='GCNConv',
help='convolution type for pooling')
parser.add_argument('--label_smoothing', type=float, default=0.1, metavar='S', help='label smoothing')
args = parser.parse_args()
dataset = zdswDataset(root='./dataset', dataname='bus_vol_mag_phase')
dataset = dataset.shuffle()
batch_size = args.batch_size
num_samples = 6800
train_dataset = dataset[:int(0.7*num_samples)]
trainloader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
val_dataset = dataset[int(0.7*num_samples):int(0.9*num_samples)]
valloader = DataLoader(val_dataset, batch_size=batch_size, shuffle=True)
test_dataset = dataset[int(0.9*num_samples):]
# test_dataset = dataset[:]
testloader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False)
train_size = int(0.7*num_samples)
d = train_dataset
class Net(torch.nn.Module):
def __init__(self, args):
super(Net, self).__init__()
self.args = args
self.pooling_ratio = args.pooling_ratio
self.nhid = args.nhid
self.dropout_ratio = args.dropout_ratio
self.conv1 = GCNConv(dataset.num_features, self.nhid)
self.pool1 = SAGPooling(self.nhid, ratio=self.pooling_ratio)
self.bn1 = nn.BatchNorm1d(self.nhid, eps=1e-05, momentum=0.1, affine=True)
self.conv2 = GCNConv(self.nhid, self.nhid)
self.pool2 = SAGPooling(self.nhid, ratio=self.pooling_ratio)
self.conv3 = GCNConv(self.nhid, self.nhid)
self.pool3 = SAGPooling(self.nhid, ratio=self.pooling_ratio)
self.lin1 = nn.Linear(self.nhid, self.nhid)
self.lin2 = nn.Linear(self.nhid, self.nhid//2)
self.lin3 = nn.Linear(self.nhid//2, dataset.num_classes)
# self.conv1 = GATConv(dataset.num_features, 128, heads=5, concat=True, dropout=0.5)
# self.conv2 = GATConv(5 * 128, 64, heads=4, concat=False, dropout=0.5)
# self.fc = nn.Sequential(
# nn.Linear(64, 256),
# nn.ReLU(),
# nn.Linear(256, 512),
# nn.ReLU(),
# nn.Linear(512, dataset.num_classes)
# )
def forward(self, data):
# x, edge_index, batch, edge_weight = data.x, data.edge_index, data.batch, data.edge_weight
x, edge_index, batch = data.x, data.edge_index, data.batch
# x = F.relu(self.conv1(x, edge_index))
# x, edge_index, _, batch, _, _ = self.pool1(x, edge_index, None, batch)
# x1 = torch.cat([gmp(x, batch), gap(x, batch)], dim=1)
#
# x = F.relu(self.conv2(x, edge_index))
# x, edge_index, _, batch, _, _ = self.pool2(x, edge_index, None, batch)
# x2 = torch.cat([gmp(x, batch), gap(x, batch)], dim=1)
#
# x = F.relu(self.conv3(x, edge_index))
# x, edge_index, _, batch, _, _ = self.pool3(x, edge_index, None, batch)
# x3 = torch.cat([gmp(x, batch), gap(x, batch)], dim=1)
#
# x = x1 + x2 + x3
# x = scatter_mean(x, data.batch, dim=0)
x = F.relu(self.conv1(x, edge_index))
x = F.relu(self.conv2(x, edge_index))
x = F.relu(self.conv3(x, edge_index))
x = self.bn1(x)
x, _ = scatter_max(x, data.batch, dim=0)
x = F.relu(self.lin1(x))
x = F.dropout(x, p=self.dropout_ratio, training=self.training)
x = F.relu(self.lin2(x))
x = F.dropout(x, p=self.dropout_ratio, training=self.training)
x = self.lin3(x)
# x = F.log_softmax(self.lin3(x), dim=-1)
return x
torch.set_num_threads(1)
# criterion = nn.CrossEntropyLoss()
criterion = LabelSmoothingCrossEntropy(smoothing=0.1)
ECE = ECELoss(n_bins=10)
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
# model = nn.DataParallel(Net, device_ids=[0, 2])
# model = model.cuda()
model = Net(args).to(device)
# optimizer = torch.optim.Adam(model.parameters(), lr=args.lr, weight_decay=args.weight_decay)
optimizer = torch.optim.SGD(model.parameters(), lr=args.lr, weight_decay=args.weight_decay,
momentum=args.momentum)
scheduler = torch.optim.lr_scheduler.ExponentialLR(optimizer, gamma=0.95, last_epoch=-1)
def train():
if epoch == 10:
for param_group in optimizer.param_groups:
param_group['lr'] = 0.0005
if epoch == 100:
for param_group in optimizer.param_groups:
param_group['lr'] = 0.0002
if epoch == 200:
for param_group in optimizer.param_groups:
param_group['lr'] = 0.0001
if epoch == 300:
for param_group in optimizer.param_groups:
param_group['lr'] = 1e-5
model.train()
loss_all = 0
best_loss = 100
print('Epoch:{}'.format(epoch))
for data in trainloader:
# data = nn.DataParallel(data, device_ids=[0, 2])
data = data.to(device)
optimizer.zero_grad()
out = model(data)
# smooth_label = smooth_one_hot(data.y, classes=3, smoothing=0.1)
# print(smooth_label)
# train_loss = F.nll_loss(out, data.y)
train_loss = criterion(out, data.y)
loss_all += train_loss.item()
train_loss.backward()
optimizer.step()
if train_loss.item() < best_loss:
best_loss = train_loss.item()
# best_label = data.y
train_avg_loss = loss_all / len(trainloader)
print("Training average loss: {:.4f}, Training best loss: {:.4f}".format(train_avg_loss, best_loss))
# print(best_label)
return train_avg_loss, best_loss
def test(loader):
model.eval()
correct = 0.
loss = 0.
topk = 0.
# eces = 0.
# avg_acc = 0.
# avg_conf = 0.
with torch.no_grad():
for data in loader:
data = data.to(device)
test_labels = data.y
out = model(data)
# ece, conf, acc = ECE(out, data.y)
# eces += ece
# avg_conf += conf
# avg_acc += acc
# target = smooth_one_hot(data.y, classes=3, smoothing=0.1)
pred = out.max(1)[1]
_, maxk = torch.topk(out, k=2, dim=-1)
test_labels = test_labels.view(-1, 1)
topk += (test_labels == maxk).sum().item()
correct += pred.eq(data.y).sum().item()
# loss += F.nll_loss(out, data.y, reduction='sum').item()
loss += criterion(out, data.y).item()
# print(avg_conf / len(loader), avg_acc / len(loader))
return correct / len(loader.dataset), topk / len(loader.dataset), loss / len(loader)
loss_list = []
acc_list = []
top2_acc_list = []
train_loss = []
best_loss = []
for epoch in range(1, args.epochs+1):
start_time = time.time()
training_loss, b_loss = train()
train_loss.append(training_loss)
best_loss.append(b_loss)
val_acc, top2_acc, val_loss = test(valloader)
lr = optimizer.param_groups[0]['lr']
# lr = scheduler.optimizer.param_groups[0]['lr']
# scheduler.step()
loss_list.append(val_loss)
acc_list.append(100. * val_acc)
top2_acc_list.append(100. * top2_acc)
end_time = time.time()
print("Validation loss:{:.4f}\taccuracy:{:.4f}\ttop2_acc:{:.4f}\tlr:{:.3e}\tTime:{:.2f}".format(val_loss, val_acc, top2_acc,
lr, end_time-start_time))
sio.savemat('val_loss.mat', mdict={'val_loss': loss_list})
sio.savemat('val_acc.mat', mdict={'val_acc': acc_list})
sio.savemat('top2_acc.mat', mdict={'top2_acc': top2_acc_list})
sio.savemat('train_loss.mat', mdict={'train_loss': train_loss})
sio.savemat('best_loss.mat', mdict={'best_loss': best_loss})
x1 = range(1, args.epochs+1)
x2 = range(1, args.epochs+1)
y1 = acc_list
y2 = loss_list
plt.subplot(2, 1, 1)
plt.plot(x1, y1, '.-')
# plt.title('Test accuracy vs. epoches')
plt.xlabel('Epochs')
plt.ylabel('Test accuracy')
plt.subplot(2, 1, 2)
plt.plot(x2, y2, '.-')
plt.xlabel('Epochs')
plt.ylabel('Loss')
plt.savefig("accuracy_loss.jpg")
plt.show()
test_acc, top2_acc, test_loss = test(testloader)
print('Epoch: {:02d}, Test: {:.4f}, Top2_Acc: {:.4f}'.format(epoch, test_acc, top2_acc))
torch.save(model.state_dict(), 'zdsw_model.pkl')
# 加载保存的模型进行测试
# activation = {}
#
#
# def get_activation(name):
# def hook(model, input, output):
# # 如果你想feature的梯度能反向传播，那么去掉 detach（）
# activation[name] = output.detach()
# return hook
#
#
# model.load_state_dict(torch.load('./result/LS alpha=0.05/zdsw_model.pkl'))
# model.lin2.register_forward_hook(get_activation('lin2'))
# val_acc, top2_acc, val_loss = test(testloader)
# print("Validation loss:{:.4f}\taccuracy:{:.4f}\ttop2_acc:{:.4f}".format(val_loss, val_acc, top2_acc))
# print(activation['lin2'].size())