-
Notifications
You must be signed in to change notification settings - Fork 0
/
Alexnet.py
131 lines (106 loc) · 6.41 KB
/
Alexnet.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
import tensorflow as tf
from fathom.imagenet import imagenet
from fathom.nn import default_runstep
#定义一个卷积层
def conv2d(name, l_input, w, b):
return tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(l_input, w, strides=[1, 1, 1, 1], padding='SAME'),b), name=name) #为什么这样定义我看不懂
def max_pool(name, l_input, k): #定义一个最大池化层函数
return tf.nn.max_pool(l_input, ksize=[1, k, k, 1], strides=[1, k, k, 1], padding='SAME', name=name) #这应该是设置池化层的扫描方法，但不大懂意思
def norm(name, l_input, lsize=4):
return tf.nn.lrn(l_input, lsize, bias=1.0, alpha=0.001 / 9.0, beta=0.75, name=name) #这个不知道代表什么意思
class AlexNet(imagenet.ImagenetModel): #定义一个类叫AlexNet，这个类从模块imagenet.ImagenetModel中继承
"""Based on Aymeric Damien's TensorFlow example of AlexNet."""
def build_inference(self, images): #构造一个预测图像的函数
with self.G.as_default():
# conv1
with tf.name_scope('conv1') as scope:
kernel = tf.Variable(tf.truncated_normal([11, 11, 3, 64], dtype=tf.float32,
stddev=1e-1), name='weights')
conv = tf.nn.conv2d(images, kernel, [1, 4, 4, 1], padding='SAME')
biases = tf.Variable(tf.constant(0.0, shape=[64], dtype=tf.float32),
trainable=True, name='biases')
bias = tf.nn.bias_add(conv, biases)
conv1 = tf.nn.relu(bias, name=scope) #这段代码应该是用滤波器扫描接受到的图像，然后得到激活图像的过程，不过代码我不大看得懂
# pool1
pool1 = tf.nn.max_pool(conv1,
ksize=[1, 3, 3, 1],
strides=[1, 2, 2, 1],
padding='VALID',
name='pool1') #这段代码就是用最大池化层处理第一个卷积层得到的图像，但是我还是不懂每一行的代码
# TODO: lrn1
lsize = 4 #不懂
norm1 = tf.nn.lrn(pool1, lsize, bias=1.0, alpha=0.001 / 9.0, beta=0.75) #这应该是一个计算函数
# conv2
with tf.name_scope('conv2') as scope:
kernel = tf.Variable(tf.truncated_normal([5, 5, 64, 192], dtype=tf.float32,stddev=1e-1), name='weights')
conv = tf.nn.conv2d(norm1, kernel, [1, 1, 1, 1], padding='SAME')
biases = tf.Variable(tf.constant(0.0, shape=[192], dtype=tf.float32),
trainable=True, name='biases')
bias = tf.nn.bias_add(conv, biases)
conv2 = tf.nn.relu(bias, name=scope) #设置第二个卷积层
# pool2
pool2 = tf.nn.max_pool(conv2,
ksize=[1, 3, 3, 1],
strides=[1, 2, 2, 1],
padding='VALID',
name='pool2') #第二个池化层
norm2 = tf.nn.lrn(pool2, lsize, bias=1.0, alpha=0.001 / 9.0, beta=0.75) #这应该是代价函数
# conv3
with tf.name_scope('conv3') as scope:
kernel = tf.Variable(tf.truncated_normal([3, 3, 192, 384],dtype=tf.float32,stddev=1e-1), name='weights')
conv = tf.nn.conv2d(norm2, kernel, [1, 1, 1, 1], padding='SAME')
biases = tf.Variable(tf.constant(0.0, shape=[384], dtype=tf.float32),trainable=True, name='biases')
bias = tf.nn.bias_add(conv, biases)
conv3 = tf.nn.relu(bias, name=scope) #第三个卷积层
# conv4
with tf.name_scope('conv4') as scope:
kernel = tf.Variable(tf.truncated_normal([3, 3, 384, 256],dtype=tf.float32,stddev=1e-1), name='weights')
conv = tf.nn.conv2d(conv3, kernel, [1, 1, 1, 1], padding='SAME')
biases = tf.Variable(tf.constant(0.0, shape=[256], dtype=tf.float32),trainable=True, name='biases')
bias = tf.nn.bias_add(conv, biases)
conv4 = tf.nn.relu(bias, name=scope) #第四个卷积层
# conv5
with tf.name_scope('conv5') as scope:
kernel = tf.Variable(tf.truncated_normal([3, 3, 256, 256],dtype=tf.float32,stddev=1e-1), name='weights')
conv = tf.nn.conv2d(conv4, kernel, [1, 1, 1, 1], padding='SAME')
biases = tf.Variable(tf.constant(0.0, shape=[256], dtype=tf.float32),trainable=True, name='biases')
bias = tf.nn.bias_add(conv, biases)
conv5 = tf.nn.relu(bias, name=scope) #第五个卷积层
# pool5
pool5 = tf.nn.max_pool(conv5,
ksize=[1, 3, 3, 1],
strides=[1, 2, 2, 1],
padding='VALID',
name='pool5') #第五个卷积层的最大池化层
pool5_shape = pool5.get_shape().as_list() #以元组的形式得到第五个池化层之后的尺寸
pool5_length = pool5_shape[1] * pool5_shape[2] * pool5_shape[3] #计算经过第五个池化层以后含有多少个变量
#这段应该是计算分类，代码看不明白
wd1 = tf.Variable(tf.random_normal([pool5_length, 4096]))
bd1 = tf.Variable(tf.random_normal([4096]))
flattened_pool5 = tf.reshape(pool5, [self.batch_size, pool5_length])
dense1 = tf.nn.relu(tf.nn.xw_plus_b(flattened_pool5, wd1, bd1), name='fc1')
wd2 = tf.Variable(tf.random_normal([4096, 4096]))
bd2 = tf.Variable(tf.random_normal([4096]))
dense2 = tf.nn.relu(tf.nn.xw_plus_b(dense1, wd2, bd2), name='fc2')
w_out = tf.Variable(tf.random_normal([4096, self.n_classes]))
b_out = tf.Variable(tf.random_normal([self.n_classes]))
self.logits = tf.nn.xw_plus_b(dense2, w_out, b_out)
return self.logits
def build_hyperparameters(self): #构造假设函数
self.learning_rate = 0.001 #学习率为0.001
self.training_iters = 200000 #训练样本数为20000
self.batch_size = 64 #批量大小为64
if self.init_options: #if这一段应该是全连接层，但是代码我看不懂
self.batch_size = self.init_options.get('batch_size', self.batch_size)
self.display_step = 1
self.dropout = 0.8 # Dropout, probability to keep units
# TODO: can this not be a placeholder?
self.keep_prob = tf.placeholder(tf.float32) # dropout (keep probability)
class AlexNetFwd(AlexNet): #定义一个AlexNet的子类
forward_only = True
#看不明白
if __name__=='__main__':
m = AlexNet()
m.setup()
m.run(runstep=default_runstep, n_steps=10)
m.teardown()