-
Notifications
You must be signed in to change notification settings - Fork 0
/
MobilePredictor.py
174 lines (148 loc) · 6.03 KB
/
MobilePredictor.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
import torch
import torch.nn as nn
from NeuralModels import SILU, Perceptron
from DeepImagePrediction import IMAGE_SIZE, DIMENSION, CHANNELS
import math
def conv_bn(inp, oup, stride , activation = nn.ReLU()):
return nn.Sequential(
nn.Conv2d(inp, oup, 3, stride, 1, bias=False),
nn.BatchNorm2d(oup),
activation,
)
def conv_1x1_bn(inp, oup, activation = nn.ReLU()):
return nn.Sequential(
nn.Conv2d(inp, oup, 1, 1, 0, bias=False),
nn.BatchNorm2d(oup),
activation,
)
class InvertedResidual(nn.Module):
def __init__(self, inp, oup, stride, expand_ratio, activation = nn.ReLU()):
super(InvertedResidual, self).__init__()
self.stride = stride
assert stride in [1, 2]
self.use_res_connect = self.stride == 1 and inp == oup
self.conv = nn.Sequential(
nn.Conv2d(inp, inp * expand_ratio, 1, 1, 0, bias=False),
nn.BatchNorm2d(inp * expand_ratio),
activation,
nn.Conv2d(inp * expand_ratio, inp * expand_ratio, 3, stride, 1, groups=inp * expand_ratio, bias=False),
nn.BatchNorm2d(inp * expand_ratio),
activation,
nn.Conv2d(inp * expand_ratio, oup, 1, 1, 0, bias=False),
nn.BatchNorm2d(oup),
)
def forward(self, x):
if self.use_res_connect:
return x + self.conv(x)
else:
return self.conv(x)
class MobileNetV2(nn.Module):
def __init__(self, dimension=1000, channels = 3, input_size=224, width_mult=1.):
super(MobileNetV2, self).__init__()
# setting of inverted residual blocks
self.interverted_residual_setting = [
# t, c, n, s
[1, 16, 1, 1],
[6, 24, 2, 2],
[6, 32, 3, 2],
[6, 64, 4, 2],
[6, 96, 3, 1],
[6, 160, 3, 2],
[6, 320, 1, 1],
]
# building first layer
assert input_size % 32 == 0
input_channel = int(32 * width_mult)
self.last_channel = int(1280 * width_mult) if width_mult > 1.0 else 1280
self.features = [conv_bn(channels, input_channel, 2)]
# building inverted residual blocks
for t, c, n, s in self.interverted_residual_setting:
output_channel = int(c * width_mult)
for i in range(n):
if i == 0:
self.features.append(InvertedResidual(input_channel, output_channel, s, t))
else:
self.features.append(InvertedResidual(input_channel, output_channel, 1, t))
input_channel = output_channel
# building last several layers
self.features.append(conv_1x1_bn(input_channel, self.last_channel))
self.features.append(nn.AvgPool2d(input_size // 32))
# make it nn.Sequential
self.features = nn.Sequential(*self.features)
# building classifier
self.classifier = nn.Sequential(
nn.Dropout(),
nn.Linear(self.last_channel, dimension),
)
self._initialize_weights()
def forward(self, x):
x = self.features(x)
x = x.view(-1, self.last_channel)
x = self.classifier(x)
return x
def _initialize_weights(self):
for m in self.modules():
if isinstance(m, nn.Conv2d):
n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
m.weight.data.normal_(0, math.sqrt(2. / n))
if m.bias is not None:
m.bias.data.zero_()
elif isinstance(m, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
elif isinstance(m, nn.Linear):
n = m.weight.size(1)
m.weight.data.normal_(0, 0.01)
m.bias.data.zero_()
MOBILE_NET_V2_UTR = 'https://s3-us-west-1.amazonaws.com/models-nima/mobilenetv2.pth.tar'
import requests
import os
def download_file(url, local_filename, chunk_size=1024):
if os.path.exists(local_filename):
return local_filename
r = requests.get(url, stream=True)
with open(local_filename, 'wb') as f:
for chunk in r.iter_content(chunk_size=chunk_size):
if chunk:
f.write(chunk)
return local_filename
def mobile_net_v2(dimension=1000, channels = 3, input_size=224, width_mult=1., pretrained = True):
model = MobileNetV2(input_size=input_size)
if pretrained:
path_to_model = './mobilenetv2.pth.tar'
if not os.path.exists(path_to_model):
path_to_model = download_file(MOBILE_NET_V2_UTR, path_to_model)
state_dict = torch.load(path_to_model, map_location=lambda storage, loc: storage)
model.load_state_dict(state_dict)
return model
class MobilePredictor(nn.Module):
def __init__(self, activation=SILU(), pretrained=True):
super(MobilePredictor, self).__init__()
self.activation = activation
base_model = mobile_net_v2(dimension=DIMENSION, channels = CHANNELS, input_size=IMAGE_SIZE, width_mult=1., pretrained = pretrained)
base_model = nn.Sequential(*list(base_model.children())[:-1])
conv = nn.Conv2d(CHANNELS, 32, kernel_size=3, stride=2, padding=3, bias=False)
weight = torch.FloatTensor(32, CHANNELS, 3, 3)
parameters = list(base_model.parameters())
for i in range(32):
if CHANNELS == 1:
weight[i, :, :, :] = parameters[0].data[i].mean(0)
else:
weight[i, :, :, :] = parameters[0].data[i]
conv.weight.data.copy_(weight)
for m in base_model.modules():
if isinstance(m, InvertedResidual):
m.conv[2] = activation
m.conv[5] = activation
self.features =base_model
self.features[0][0]= conv
self.predictor = nn.Sequential(
Perceptron(1280, 1280),
nn.Dropout(p=0),
activation,
Perceptron( 1280, DIMENSION),
)
def forward(self, x):
x = self.features(x)
x = self.predictor(x)
return torch.sigmoid(x)