dissertation/_cnn.py at main · HelloYuiYui/dissertation · GitHub

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
import torch
import torch.nn as nn
import torch.nn.functional as F
import os
import sys
import torch.autograd as autograd
import torch.nn.functional as F


class CNN(nn.Module):

    def __init__(self, args):
        super(CNN, self).__init__()
        self.args = args

        V = args.embed_num
        D = args.embed_dim
        C = args.class_num
        Ci = 1
        Co = args.kernel_num
        Ks = args.kernel_sizes

        self.embed = nn.Embedding(V, D)
        self.convs = nn.ModuleList([nn.Conv2d(Ci, Co, (K, D)) for K in Ks])
        self.dropout = nn.Dropout(args.dropout)
        self.fc1 = nn.Linear(len(Ks) * Co, C)

        if self.args.static:
            self.embed.weight.requires_grad = False

    def forward(self, x):
        x = self.embed(x)  # (N, W, D)

        x = x.unsqueeze(1)  # (N, Ci, W, D)

        x = [F.relu(conv(x)).squeeze(3) for conv in self.convs]  # [(N, Co, W), ...]*len(Ks)

        x = [F.max_pool1d(i, i.size(2)).squeeze(2) for i in x]  # [(N, Co), ...]*len(Ks)

        x = torch.cat(x, 1)

        x = self.dropout(x)  # (N, len(Ks)*Co)
        logit = self.fc1(x)  # (N, C)
        return logit

    def train(self, train_iter, dev_iter, model, args):
        if args.cuda:
            self.cuda()

        optimizer = torch.optim.Adam(model.parameters(), lr=args.lr)

        steps = 0
        best_acc = 0
        last_step = 0
        for epoch in range(1, args.epochs+1):
            for batch in train_iter:
                model.train()
                feature, target = batch.text, batch.label
                feature.t_(), target.sub_(1)  # batch first, index align
                if args.cuda:
                    feature, target = feature.cuda(), target.cuda()

                optimizer.zero_grad()
                logit = model(feature)
                loss = F.cross_entropy(logit, target)
                loss.backward()
                optimizer.step()

                steps += 1
                if steps % args.log_interval == 0:
                    corrects = (torch.max(logit, 1)[1].view(target.size()).data == target.data).sum()
                    accuracy = 100.0 * corrects/batch.batch_size
                    sys.stdout.write(
                        '\rBatch[{}] - loss: {:.6f}  acc: {:.4f}%({}/{})'.format(steps,
                                                                                loss.item(),
                                                                                accuracy.item(),
                                                                                corrects.item(),
                                                                                batch.batch_size))
                if steps % args.test_interval == 0:
                    dev_acc = eval(dev_iter, model, args)
                    if dev_acc > best_acc:
                        best_acc = dev_acc
                        last_step = steps
                        if args.save_best:
                            self.save(model, args.save_dir, 'best', steps)
                    else:
                        if steps - last_step >= args.early_stop:
                            print('early stop by {} steps.'.format(args.early_stop))
                elif steps % args.save_interval == 0:
                    self.save(model, args.save_dir, 'snapshot', steps)


    def eval(data_iter, model, args):
        model.eval()
        corrects, avg_loss = 0, 0
        for batch in data_iter:
            feature, target = batch.text, batch.label
            feature.t_(), target.sub_(1)  # batch first, index align
            if args.cuda:
                feature, target = feature.cuda(), target.cuda()

            logit = model(feature)
            loss = F.cross_entropy(logit, target, size_average=False)

            avg_loss += loss.item()
            corrects += (torch.max(logit, 1)
                        [1].view(target.size()).data == target.data).sum()

        size = len(data_iter.dataset)
        avg_loss /= size
        accuracy = 100.0 * corrects/size
        print('\nEvaluation - loss: {:.6f}  acc: {:.4f}%({}/{}) \n'.format(avg_loss,
                                                                        accuracy,
                                                                        corrects,
                                                                        size))
        return accuracy


    def predict(self, text, text_field, label_feild, cuda_flag):
        assert isinstance(text, str)
        self.eval()
        # text = text_field.tokenize(text)
        text = text_field.preprocess(text)
        text = [[text_field.vocab.stoi[x] for x in text]]
        x = torch.tensor(text)
        x = autograd.Variable(x)
        if cuda_flag:
            x = x.cuda()
        print(x)
        output = self(x)
        _, predicted = torch.max(output, 1)
        return label_feild.vocab.itos[predicted.item()+1]


    def save(model, save_dir, save_prefix, steps):
        if not os.path.isdir(save_dir):
            os.makedirs(save_dir)
        save_prefix = os.path.join(save_dir, save_prefix)
        save_path = '{}_steps_{}.pt'.format(save_prefix, steps)
        torch.save(model.state_dict(), save_path)