tensorflow实现残差网络（mnist数据集）

介绍


残差网络是何凯明大神的神作，效果非常好，深度可以达到1000层。但是，其实现起来并没有那末难，在这里以tensorflow作为框架，实现基于mnist数据集上的残差网络，当然只是比较浅层的。

如下图所示：

*
实线的Connection部分，表示通道相同，如上图的第一个粉色矩形和第三个粉色矩形，都是3x3x64的特征图，由于通道相同，所以采用计算方式为H(x)=F(x)+x
*
虚线的的Connection部分，表示通道不同，如上图的第一个绿色矩形和第三个绿色矩形，分别是3x3x64和3x3x128的特征图，通道不同，采用的计算方式为H(x)=F(x)+Wx，其中W是卷积操作，用来调整x维度的。






根据输入和输出尺寸是否相同，又分为identity_block和conv_block，每种block有上图两种模式，三卷积和二卷积，三卷积速度更快些，因此在这里选择该种方式。具体实现见如下代码：
#tensorflow基于mnist数据集上的VGG11网络，可以直接运行 from tensorflow.examples.tutorials.mnist
import input_data import tensorflow as tf #tensorflow基于mnist实现VGG11 mnist =
input_data.read_data_sets('MNIST_data', one_hot=True) #x=mnist.train.images
#y=mnist.train.labels #X=mnist.test.images #Y=mnist.test.labels x =
tf.placeholder(tf.float32, [None,784]) y = tf.placeholder(tf.float32, [None, 10
]) sess = tf.InteractiveSession()def weight_variable(shape): #这里是构建初始变量 initial
= tf.truncated_normal(shape, mean=0,stddev=0.1) #创建变量 return
tf.Variable(initial)def bias_variable(shape): initial = tf.constant(0.1,
shape=shape)return tf.Variable(initial) #在这里定义残差网络的id_block块，此时输入和输出维度相同 def
identity_block(X_input, kernel_size, in_filter, out_filters, stage, block): """
Implementation of the identity block as defined in Figure 3 Arguments: X --
input tensor of shape (m, n_H_prev, n_W_prev, n_C_prev) kernel_size -- integer,
specifying the shape of the middle CONV's window for the main path filters --
python list of integers, defining the number of filters in the CONV layers of
the main path stage -- integer, used to name the layers, depending on their
position in the network block -- string/character, used to name the layers,
depending on their position in the network training -- train or test Returns: X
-- output of the identity block, tensor of shape (n_H, n_W, n_C) """ # defining
name basis block_name = 'res' + str(stage) + block f1, f2, f3 = out_filters with
tf.variable_scope(block_name): X_shortcut = X_input#first W_conv1 =
weight_variable([1, 1, in_filter, f1]) X = tf.nn.conv2d(X_input, W_conv1,
strides=[1, 1, 1, 1], padding='SAME') b_conv1 = bias_variable([f1]) X =
tf.nn.relu(X+ b_conv1)#second W_conv2 = weight_variable([kernel_size,
kernel_size, f1, f2]) X = tf.nn.conv2d(X, W_conv2, strides=[1, 1, 1, 1],
padding='SAME') b_conv2 = bias_variable([f2]) X = tf.nn.relu(X+ b_conv2) #third
W_conv3 = weight_variable([1, 1, f2, f3]) X = tf.nn.conv2d(X, W_conv3, strides=[
1, 1, 1, 1], padding='SAME') b_conv3 = bias_variable([f3]) X = tf.nn.relu(X+
b_conv3)#final step add = tf.add(X, X_shortcut) b_conv_fin =
bias_variable([f3]) add_result = tf.nn.relu(add+b_conv_fin)return add_result
#这里定义conv_block模块，由于该模块定义时输入和输出尺度不同，故需要进行卷积操作来改变尺度，从而得以相加 def
convolutional_block( X_input, kernel_size, in_filter, out_filters, stage,
block, stride=2): """ Implementation of the convolutional block as defined in
Figure 4 Arguments: X -- input tensor of shape (m, n_H_prev, n_W_prev,
n_C_prev) kernel_size -- integer, specifying the shape of the middle CONV's
window for the main path filters -- python list of integers, defining the
number of filters in the CONV layers of the main path stage -- integer, used to
name the layers, depending on their position in the network block --
string/character, used to name the layers, depending on their position in the
network training -- train or test stride -- Integer, specifying the stride to
be used Returns: X -- output of the convolutional block, tensor of shape (n_H,
n_W, n_C) """ # defining name basis block_name = 'res' + str(stage) + block with
tf.variable_scope(block_name): f1, f2, f3 = out_filters x_shortcut = X_input
#first W_conv1 = weight_variable([1, 1, in_filter, f1]) X =
tf.nn.conv2d(X_input, W_conv1,strides=[1, stride, stride, 1],padding='SAME')
b_conv1 = bias_variable([f1]) X = tf.nn.relu(X + b_conv1)#second W_conv2
=weight_variable([kernel_size, kernel_size, f1, f2]) X = tf.nn.conv2d(X,
W_conv2, strides=[1,1,1,1], padding='SAME') b_conv2 = bias_variable([f2]) X =
tf.nn.relu(X+b_conv2)#third W_conv3 = weight_variable([1,1, f2,f3]) X =
tf.nn.conv2d(X, W_conv3, strides=[1, 1, 1,1], padding='SAME') b_conv3 =
bias_variable([f3]) X = tf.nn.relu(X+b_conv3)#shortcut path W_shortcut
=weight_variable([1, 1, in_filter, f3]) x_shortcut = tf.nn.conv2d(x_shortcut,
W_shortcut, strides=[1, stride, stride, 1], padding='VALID') #final add =
tf.add(x_shortcut, X)#建立最后融合的权重 b_conv_fin = bias_variable([f3]) add_result =
tf.nn.relu(add+ b_conv_fin)return add_result x = tf.reshape(x, [-1,28,28,1])
w_conv1 = weight_variable([2, 2, 1, 64]) x = tf.nn.conv2d(x, w_conv1, strides=[1
,2, 2, 1], padding='SAME') b_conv1 = bias_variable([64]) x =
tf.nn.relu(x+b_conv1)#这里操作后变成14x14x64 x = tf.nn.max_pool(x, ksize=[1, 3, 3, 1],
strides=[1, 1, 1, 1], padding='SAME') #stage 2 x =
convolutional_block(X_input=x, kernel_size=3, in_filter=64, out_filters=[64, 64,
256], stage=2, block='a', stride=1) #上述conv_block操作后，尺寸变为14x14x256 x =
identity_block(x,3, 256, [64, 64, 256], stage=2, block='b' ) x =
identity_block(x,3, 256, [64, 64, 256], stage=2, block='c')
#上述操作后张量尺寸变成14x14x256 x = tf.nn.max_pool(x, [1, 2, 2, 1], strides=[1,2,2,1],
padding='SAME') #变成7x7x256 flat = tf.reshape(x, [-1,7*7*256]) w_fc1 =
weight_variable([7 * 7 *256, 1024]) b_fc1 = bias_variable([1024]) h_fc1 =
tf.nn.relu(tf.matmul(flat, w_fc1) + b_fc1) keep_prob =
tf.placeholder(tf.float32) h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob) w_fc2 =
weight_variable([1024, 10]) b_fc2 = bias_variable([10]) y_conv =
tf.matmul(h_fc1_drop, w_fc2) + b_fc2#建立损失函数，在这里采用交叉熵函数 cross_entropy =
tf.reduce_mean( tf.nn.softmax_cross_entropy_with_logits(labels=y,
logits=y_conv)) train_step = tf.train.AdamOptimizer(1e-3
).minimize(cross_entropy) correct_prediction = tf.equal(tf.argmax(y_conv,1),
tf.argmax(y,1)) accuracy = tf.reduce_mean(tf.cast(correct_prediction,
tf.float32))#初始化变量 sess.run(tf.global_variables_initializer()) print("cuiwei")
for i in range(2000): batch = mnist.train.next_batch(10) if i%100 == 0:
train_accuracy = accuracy.eval(feed_dict={ x:batch[0], y: batch[1], keep_prob:
1.0}) print("step %d, training accuracy %g"%(i, train_accuracy))
train_step.run(feed_dict={x: batch[0], y: batch[1], keep_prob: 0.5})

• « 上一篇：Dubbo实战入门，良心详解之作
• » 下一篇：IDEA 调试js代码