DAM-Net/training_script.py at main · ami-lab-knut/DAM-Net · 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
141
142
143
144
145
146
147
148
from model import DAM_NET
from data_loaders import DataPrep, ValDataPrep
from torch.utils.data import DataLoader, Dataset, random_split
import torch.optim as optim
import torch
import torch.nn as nn
import numpy as np
from sklearn.metrics import accuracy_score, precision_score
from sklearn.metrics import confusion_matrix
from utils import scores


device = torch.device('cuda:0')

model = DAM_NET(3, 3).to(device)

epochs = 50
base_lr = 0.001
optimizer = torch.optim.Adam(model.parameters(), lr = base_lr, weight_decay = 0.0001)


loss_function = nn.CrossEntropyLoss(label_smoothing = 0.2)

data_path = 'E:\\Research\\covid_research\\total_covid_data_3classes.npy'
training_set = DataPrep(data_path)
validation_set = ValDataPrep(data_path)

train_loader = DataLoader(training_set, batch_size=12, shuffle=True, pin_memory=True)
val_loader = DataLoader(validation_set, batch_size=12, shuffle=True, pin_memory=True)

for epoch in range(epochs):
    train_accuracy = []
    train_sensitivity = []
    train_specivity = []
    train_precision = []

    train_accuracy_1 = []
    train_precision_1 = []
    train_sensitivity_1 = []
    train_specificity_1 = []

    train_accuracy_2 = []
    train_precision_2 = []
    train_sensitivity_2 = []
    train_specificity_2 = []

    for sample in train_loader:
        image, label = sample
        image, label = image.to(device), label.to(device)
        label = torch.squeeze(label, dim = 1)

        output = model(image.float())

        label = torch.squeeze(label, dim = 1)
        loss = loss_function(output, label)

        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        output = torch.argmax(output, dim = 1)
        label = torch.argmax(label, dim = 1)

        label = label.detach().cpu().numpy()
        output = output.detach().cpu().numpy()

        cm = confusion_matrix(label, output)
        accuracy_1, precision_1, sensitivity_1, specificity_1, accuracy_2, precision_2, sensitivity_2, specificity_2 = scores(cm)
        train_accuracy_1.append(accuracy_1)
        train_precision_1.append(precision_1)
        train_sensitivity_1.append(sensitivity_1)
        train_specificity_1.append(specificity_1)

        train_accuracy_2.append(accuracy_2)
        train_precision_2.append(precision_2)
        train_sensitivity_2.append(sensitivity_2)
        train_specificity_2.append(specificity_2)


    all_acc = (np.mean(train_accuracy_1) + np.mean(train_accuracy_2)) / 2
    all_prec = (np.mean(train_precision_1) + np.mean(train_precision_2)) / 2
    all_sens = (np.mean(train_sensitivity_1) + np.mean(train_sensitivity_2)) / 2
    all_spec = (np.mean(train_specificity_1) + np.mean(train_specificity_2)) / 2

    print('Training Accuracy: ', all_acc)
    print('Training Precision: ', all_prec)
    print('Training Sensitivity: ', all_sens)
    print('Training Specivity: ', all_spec)


    val_accuracy = []
    val_sensitivity = []
    val_specivity = []
    val_precision = []

    val_accuracy_1 = []
    val_precision_1 = []
    val_sensitivity_1 = []
    val_specificity_1 = []

    val_accuracy_2 = []
    val_precision_2 = []
    val_sensitivity_2 = []
    val_specificity_2 = []

    for sample in val_loader:
        image, label = sample
        image, label = image.to(device), label.to(device)

        output = model(image.float())

        label = torch.squeeze(label, dim = 1)
        loss = loss_function(output, label)

        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        output = torch.argmax(output, dim = 1)
        label = torch.argmax(label, dim = 1)

        label = label.detach().cpu().numpy()
        output = output.detach().cpu().numpy()

        cm = confusion_matrix(label, output)
        accuracy_1, precision_1, sensitivity_1, specificity_1, accuracy_2, precision_2, sensitivity_2, specificity_2 = scores(cm)
        val_accuracy_1.append(accuracy_1)
        val_precision_1.append(precision_1)
        val_sensitivity_1.append(sensitivity_1)
        val_specificity_1.append(specificity_1)

        val_accuracy_2.append(accuracy_2)
        val_precision_2.append(precision_2)
        val_sensitivity_2.append(sensitivity_2)
        val_specificity_2.append(specificity_2)


    all_acc = (np.mean(val_accuracy_1) + np.mean(val_accuracy_2)) / 2
    all_prec = (np.mean(val_precision_1) + np.mean(val_precision_2)) / 2
    all_sens = (np.mean(val_sensitivity_1) + np.mean(val_sensitivity_2)) / 2
    all_spec = (np.mean(val_specificity_1) + np.mean(val_specificity_2)) / 2

    print('Validation Accuracy: ', all_acc)
    print('Validation Precision: ', all_prec)
    print('Validation Sensitivity: ', all_sens)
    print('Validation Specivity: ', all_spec)

    torch.save(model.state_dict(), 'E:\\Research\\covid_research\\saved_models\\binary_only\\dam_net{}.pth'.format(epoch+1))