-
Notifications
You must be signed in to change notification settings - Fork 38
/
model.py
182 lines (144 loc) · 7.25 KB
/
model.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
#coding:utf8
import torch
import torch.nn as nn
import torchvision
from PIL import Image, ImageDraw
import numpy as np
# cfg = {
# 'VGG11': [64, 'M', 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
# 'VGG13': [64, 64, 'M', 128, 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'],
# 'VGG16': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 'M', 512, 512, 512, 'M', 512, 512, 512, 'M'],
# 'VGG19': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 256, 'M', 512, 512, 512, 512, 'M', 512, 512, 512, 512, 'M'],
# }
featur_cfg = ''
class VGG(nn.Module):
def __init__(self):
super(VGG, self).__init__()
cfg = [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 'M', 512, 512, 512, 'M', 512, 512, 512]
self.features = self._make_layers(cfg)
self._rpn_model()
size = (7, 7)
self.adaptive_max_pool = torch.nn.AdaptiveMaxPool2d(size[0], size[1])
self.roi_classifier()
def _make_layers(self, cfg):
layers = []
in_channels = 3
for x in cfg:
if x == 'M':
layers += [nn.MaxPool2d(kernel_size=2, stride=2)]
else:
layers += [nn.Conv2d(in_channels, x, kernel_size=3, padding=1),
nn.BatchNorm2d(x),
nn.ReLU(inplace=True)]
in_channels = x
# layers += [nn.Conv2d(in_channels, 512, kernel_size=3, padding=1)]
return nn.Sequential(*layers)
# return layers
def _rpn_model(self, mid_channels=512, in_channels=512, n_anchor=9):
self.rpn_conv = nn.Conv2d(in_channels, mid_channels, 3, 1, 1)
self.reg_layer = nn.Conv2d(mid_channels, n_anchor * 4, 1, 1, 0)
# I will be going to use softmax here. you can equally use sigmoid if u replace 2 with 1.
self.cls_layer = nn.Conv2d(mid_channels, n_anchor * 2, 1, 1, 0)
# conv sliding layer
self.rpn_conv.weight.data.normal_(0, 0.01)
self.rpn_conv.bias.data.zero_()
# Regression layer
self.reg_layer.weight.data.normal_(0, 0.01)
self.reg_layer.bias.data.zero_()
# classification layer
self.cls_layer.weight.data.normal_(0, 0.01)
self.cls_layer.bias.data.zero_()
def forward(self, data):
out_map = self.features(data)
# for layer in self.features:
# # print layer
# data = layer(data)
# # print data.data.shape
#
# # out = data.view(data.size(0), -1)
x = self.rpn_conv(out_map)
pred_anchor_locs = self.reg_layer(x) # 回归层，计算有效anchor转为目标框的四个系数
pred_cls_scores = self.cls_layer(x) # 分类层，判断该anchor是否可以捕获目标
return out_map, pred_anchor_locs, pred_cls_scores
def roi_classifier(self, class_num=20): # 假设为VOC数据集，共20分类
# 分类层
self.roi_head_classifier = nn.Sequential(*[nn.Linear(25088, 4096),
nn.ReLU(),
nn.Linear(4096, 4096),
nn.ReLU()])
self.cls_loc = nn.Linear(4096, (class_num+1) * 4) # (VOC 20 classes + 1 background. Each will have 4 co-ordinates)
self.cls_loc.weight.data.normal_(0, 0.01)
self.cls_loc.bias.data.zero_()
self.score = nn.Linear(4096, class_num+1) # (VOC 20 classes + 1 background)
def rpn_loss(self, rpn_loc, rpn_score, gt_rpn_loc, gt_rpn_label, weight=10.0):
# 对与classification我们使用Cross Entropy损失
gt_rpn_label = torch.autograd.Variable(gt_rpn_label.long())
rpn_cls_loss = torch.nn.functional.cross_entropy(rpn_score, gt_rpn_label, ignore_index=-1)
# print(rpn_cls_loss) # Variable containing: 0.6931
# 对于 Regression 我们使用smooth L1 损失
pos = gt_rpn_label.data > 0 # Regression 损失也被应用在有正标签的边界区域中
mask = pos.unsqueeze(1).expand_as(rpn_loc)
# print(mask.shape) # (22500L, 4L)
# 现在取有正数标签的边界区域
mask_pred_loc = rpn_loc[mask].view(-1, 4)
mask_target_loc = gt_rpn_loc[mask].view(-1, 4)
# print(mask_pred_loc.shape, mask_target_loc.shape) # ((18L, 4L), (18L, 4L))
# regression损失应用如下
x = np.abs(mask_target_loc.numpy() - mask_pred_loc.data.numpy())
# print x.shape # (18, 4)
# print (x < 1)
rpn_loc_loss = ((x < 1) * 0.5 * x ** 2) + ((x >= 1) * (x - 0.5))
# print rpn_loc_loss.shape # (18, 4)
rpn_loc_loss = rpn_loc_loss.sum() # 1.1628926242031001
# print rpn_loc_loss
# print rpn_loc_loss.shape
# rpn_loc_loss = np.squeeze(rpn_loc_loss)
# print rpn_loc_loss
N_reg = (gt_rpn_label > 0).float().sum()
N_reg = np.squeeze(N_reg.data.numpy())
# print "N_reg: {}, {}".format(N_reg, N_reg.shape)
rpn_loc_loss = rpn_loc_loss / N_reg
rpn_loc_loss = np.float32(rpn_loc_loss)
# rpn_loc_loss = torch.autograd.Variable(torch.from_numpy(rpn_loc_loss))
rpn_cls_loss = np.squeeze(rpn_cls_loss.data.numpy())
# print "rpn_cls_loss: {}".format(rpn_cls_loss) # 0.693146109581
# print 'rpn_loc_loss: {}'.format(rpn_loc_loss) # 0.0646051466465
rpn_loss = rpn_cls_loss + (weight * rpn_loc_loss)
# print("rpn_loss: {}".format(rpn_loss)) # 1.33919757605
return rpn_loss
def roi_loss(self, pre_loc, pre_conf, target_loc, target_conf, weight=10.0):
# 分类损失
target_conf = torch.autograd.Variable(target_conf.long())
pred_conf_loss = torch.nn.functional.cross_entropy(pre_conf, target_conf, ignore_index=-1)
# print(pred_conf_loss) # Variable containing: 3.0515
# 对于 Regression 我们使用smooth L1 损失
# 用计算RPN网络回归损失的方法计算回归损失
# pre_loc_loss = REGLoss(pre_loc, target_loc)
pos = target_conf.data > 0 # Regression 损失也被应用在有正标签的边界区域中
mask = pos.unsqueeze(1).expand_as(pre_loc) # (128, 4L)
# 现在取有正数标签的边界区域
mask_pred_loc = pre_loc[mask].view(-1, 4)
mask_target_loc = target_loc[mask].view(-1, 4)
# print(mask_pred_loc.shape, mask_target_loc.shape) # ((19L, 4L), (19L, 4L))
x = np.abs(mask_target_loc.numpy() - mask_pred_loc.data.numpy())
# print x.shape # (19, 4)
pre_loc_loss = ((x < 1) * 0.5 * x ** 2) + ((x >= 1) * (x - 0.5))
# print(pre_loc_loss.sum()) # 1.4645805211187053
N_reg = (target_conf > 0).float().sum()
N_reg = np.squeeze(N_reg.data.numpy())
pre_loc_loss = pre_loc_loss.sum() / N_reg
pre_loc_loss = np.float32(pre_loc_loss)
# print pre_loc_loss # 0.077294916
# pre_loc_loss = torch.autograd.Variable(torch.from_numpy(pre_loc_loss))
# 损失总和
pred_conf_loss = np.squeeze(pred_conf_loss.data.numpy())
total_loss = pred_conf_loss + (weight * pre_loc_loss)
return total_loss
if __name__ == '__main__':
vgg = VGG()
print vgg
data = torch.randn((1, 3, 800, 800))
print data.shape
data = torch.autograd.Variable(data)
out = vgg.forward(data)
print out.data.shape