tensorFlow入门实践（三）初识AlexNet实现结构

import tensorflow as tfbatch_size = 32num_batches = 100# 显示网络每一层结构，展示每一个卷积层或池化层输出tensor的尺寸,接受一个tensor作为输入def print_activations(t): ???print(t.op.name, ‘ ‘, t.get_shape().as_list())# 接受images作为输入，返回最后一层pool5（第五个池化层）# 及parameters(AlexNet中所有需要训练的模型参数)‘‘‘def inference(images): ???parameters = [] ???with tf.name_scope(‘conv1‘) as scope: ???????# 用截断的正态分布函数（标准差为0.1）初始化卷积核的参数kernel。卷积核尺寸为11*11，颜色通道为3，卷积核数量为64 ???????kernel = tf.Variable(tf.truncated_normal([11, 11, 3, 64], ???????????????????dtype=tf.float32, stddev=1e-1), name=‘weights‘) ???????# 使用tf.nn.conv2d对输入images完成卷积操作 ???????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) ???????print_activations(conv1) ???????parameters += [kernel, biases] ???# LRN处理和最大池化处理 ???lrn1 = tf.nn.lrn(conv1, 4, bias=1.0, alpha=0.001/9, beta=0.75, name=‘lrn1‘) ???pool1 = tf.nn.max_pool(lrn1, ksize=[1, 3, 3, 1], strides=[1, 2, 2, 1], ??????????????????????????padding=‘VALID‘, name=‘pool1‘) ???print_activations(pool1) # 打印输出结果pool1的结构 ???# 设计第二个卷积层 卷积核尺寸5*5 输入通道数64 卷积核数量192 ???with tf.name_scope(‘conv2‘) as scope: ???????kernel = tf.Variable(tf.truncated_normal([5, 5, 64, 192], ????????????????????????????????????????????????dthpe=tf.float32, stddev=1e-1), name=‘weights‘) ???????# 卷积步长全部设为1,即扫描全图像素 ???????conv = tf.nn.conv2d(pool1, 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) ???????parameters += [kernel, biases] ???????print_activations(conv2) ???lrn2 = tf.nn.lrn(conv2, 4, bias=1.0, alpha=0.001/9, beta=0.75, name=‘lrn2‘) ???pool2 = tf.nn.max_pool(lrn2, ksize=[1, 3, 3, 1], strides=[1, 2, 2, 1], ??????????????????????????padding=‘VALID‘, name=‘pool2‘) ???print_activations(pool2) ???# 创建第三个卷积层 卷积核尺寸3*3 输入通道数192 卷积核数量384 步长全为1 ???with tf.name_scope(‘conv3‘) as scope: ???????kernel = tf.Variable(tf.truncated_normal([3, 3, 192, 384], ????????????????????????????????????????????????dthpe=tf.float32, stddev=1e-1), name=‘weights‘) ???????# 卷积步长全部设为1,即扫描全图像素 ???????conv = tf.nn.conv2d(pool2, 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) ???????parameters += [kernel, biases] ???????print_activations(conv3) ???# 创建第四个卷积层 卷积核尺寸3*3 输入通道数384 卷积核数量降为256 ???with tf.name_scope(‘conv4‘) as scope: ???????kernel = tf.Variable(tf.truncated_normal([3, 3, 384, 256], ????????????????????????????????????????????????dthpe=tf.float32, stddev=1e-1), name=‘weights‘) ???????# 卷积步长全部设为1,即扫描全图像素 ???????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) ???????parameters += [kernel, biases] ???????print_activations(conv4) ???????# 最后的第五个卷积层 卷积核尺寸3*3 输入通道数256 卷积核数量为256 ???????with tf.name_scope(‘conv5‘) as scope: ???????????kernel = tf.Variable(tf.truncated_normal([3, 3, 256, 256], ????????????????????????????????????????????????????dthpe=tf.float32, stddev=1e-1), name=‘weights‘) ???????????# 卷积步长全部设为1,即扫描全图像素 ???????????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) ???????????parameters += [kernel, biases] ???????????print_activations(conv5) ???????# 在5个卷积层之后，还有一个最大池化层，这个池化层和前两个卷积层后的池化层一致 ???????pool5 = tf.nn.max_pool(conv5, ksize=[1, 3, 3, 1], strides=[1, 2, 2, 1], ??????????????????????????????padding=‘VALID‘, name=‘pool5‘) ???????print_activations(pool5) ???????return pool5, parameters ???????# 在正式使用AlexNet来训练或预测时，还需要添加3个全连接层，隐含节点数分别为4096、4096、1000