-
Notifications
You must be signed in to change notification settings - Fork 0
/
network.py
136 lines (96 loc) · 4.17 KB
/
network.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
import torch
import numpy as np
import time
from torch import nn
from torch import optim
import torch.nn.functional as F
from torchvision import datasets, transforms
from collections import OrderedDict
import torchvision.models as models
import timeit
import json
import loaddata
def select_model(name='vgg19'):
if name == 'vgg19':
model, feature_count = models.vgg19_bn(pretrained=True), 25088
elif name == 'densenet161':
model, feature_count = models.densenet161(pretrained=True), 2208
elif name == 'densenet121':
model, feature_count = models.densenet121(pretrained=True), 1024
else:
model, feature_count = models.alexnet(pretrained=True), 9216
return model, feature_count
def build_classifier(input_size, hidden_layer, dropout):
orderd_dict = OrderedDict([])
if hidden_layer == 0:
print("Dropout and hidden units are ignored for zero hidden layers")
else:
start = input_size
layer = 1
for i in hidden_layer:
orderd_dict.update({'fc{}'.format(layer): nn.Linear(start, i)})
orderd_dict.update({'relu{}'.format(layer): nn.ReLU()})
orderd_dict.update({'do{}'.format(layer): nn.Dropout(dropout)})
start = i
layer += 1
orderd_dict.update({'fcout': nn.Linear(start, 102)})
orderd_dict.update({'output': nn.LogSoftmax(dim=1)})
classifier = nn.Sequential(orderd_dict)
return classifier
def validation(model, testloader, criterion, architecture):
test_loss = 0
accuracy = 0
for images, labels in testloader:
images, labels = images.to(architecture), labels.to(architecture)
output = model.forward(images)
test_loss += criterion(output, labels).item()
ps = torch.exp(output)
equality = (labels.data == ps.max(dim=1)[1])
accuracy += equality.type(torch.FloatTensor).mean()
return test_loss, accuracy
def train_model(model, trainloader, testloader, criterion, optimizer, epochs=10, print_every=10, architecture='cuda'):
model.to(architecture)
steps = 0
start = time.time()
for e in range(epochs):
running_loss = 0
model.train()
for images, labels in iter(trainloader):
steps += 1
images, labels = images.to(architecture), labels.to(architecture)
optimizer.zero_grad()
# Forward and backward passes
output = model.forward(images)
loss = criterion(output, labels)
loss.backward()
optimizer.step()
running_loss += loss.item()
if steps % print_every == 0:
end = time.time()
model.eval()
with torch.no_grad():
test_loss, accuracy = validation(model, testloader, criterion, architecture)
print("Epoch: {}/{}.. ".format(e+1, epochs),
"Time needed: {:.3f}.. ".format(end - start),
"Running Loss: {:.3f}.. ".format(running_loss/print_every),
"Validation Loss: {:.3f}.. ".format(test_loss/len(testloader)),
"Validation Accuracy: {:.3f}".format(accuracy/len(testloader)))
running_loss = 0
model.train()
start = time.time()
def build_optimizer(model, lr):
criterion = nn.NLLLoss()
optimizer = optim.Adam(model.classifier.parameters(), lr)
return criterion, optimizer
def build_and_train_model(name, hidden_layer, epochs, dropout, lr, architecture, dataloader_train, dataloader_validation, dataloader_test):
model, feature_count = select_model(name)
model.to(architecture)
# freeze parameters
for param in model.parameters():
param.requires_grad = False
# replace classifier
model.classifier = build_classifier(feature_count, hidden_layer, dropout)
print(model)
criterion, optimizer = build_optimizer(model, lr)
train_model(model, dataloader_train, dataloader_validation, criterion, optimizer, epochs=epochs, architecture=architecture)
return model