使用L2正则化和平均滑动模型的LeNet-5MNIST手写数字识别模型
使用L2正则化和平均滑动模型的LeNet-5MNIST手写数字识别模型
觉得有用的话,欢迎一起讨论相互学习~
参考文献Tensorflow实战Google深度学习框架
实验平台:
Tensorflow1.4.0
python3.5.0
MNIST数据集将四个文件下载后放到当前目录下的MNIST_data文件夹下
L2正则化
Dropout
滑动平均方法
定义模型框架与前向传播
import tensorflow as tf
# 配置神经网络的参数
INPUT_NODE = 784
OUTPUT_NODE = 10
IMAGE_SIZE = 28
NUM_CHANNELS = 1
NUM_LABELS = 10
# 第一层卷积层的尺寸和深度
CONV1_DEEP = 32
CONV1_SIZE = 5
# 第二层卷积层的尺寸和深度
CONV2_DEEP = 64
CONV2_SIZE = 5
# 全连接层的节点个数
FC_SIZE = 512
# 定义卷积神经网络的前向传播过程,这里添加了一个参数train,用于区分训练过程和测试过程。
# 这里使用dropout方法,dropout方法可以进一步提升模型可靠性并防止过拟合,dropout只在训练过程中使用。
def inference(input_tensor, train, regularizer):
# 通过使用不同的命名空间来隔离变量,可以使每一层的变量命名只需要考虑在当前层的作用,而不需要考虑重名的问题
with tf.variable_scope('layer1-conv1'):
conv1_weights = tf.get_variable(
"weight", [CONV1_SIZE, CONV1_SIZE, NUM_CHANNELS, CONV1_DEEP],
initializer=tf.truncated_normal_initializer(stddev=0.1))
conv1_biases = tf.get_variable("bias", [CONV1_DEEP], initializer=tf.constant_initializer(0.0))
conv1 = tf.nn.conv2d(input_tensor, conv1_weights, strides=[1, 1, 1, 1], padding='SAME')
relu1 = tf.nn.relu(tf.nn.bias_add(conv1, conv1_biases))
with tf.name_scope("layer2-pool1"):
pool1 = tf.nn.max_pool(relu1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding="SAME")
with tf.variable_scope("layer3-conv2"):
conv2_weights = tf.get_variable(
"weight", [CONV2_SIZE, CONV2_SIZE, CONV1_DEEP, CONV2_DEEP],
initializer=tf.truncated_normal_initializer(stddev=0.1))
conv2_biases = tf.get_variable("bias", [CONV2_DEEP], initializer=tf.constant_initializer(0.0))
conv2 = tf.nn.conv2d(pool1, conv2_weights, strides=[1, 1, 1, 1], padding='SAME')
relu2 = tf.nn.relu(tf.nn.bias_add(conv2, conv2_biases))
with tf.name_scope("layer4-pool2"):
pool2 = tf.nn.max_pool(relu2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
# pool2.getshape函数可以得到第四层输出矩阵的维度而不需要手工计算。
# 注意因为每一层神经网络的输入输出都为一个batch矩阵,所以这里得到的维度也包含了一个batch中数据的个数。
pool_shape = pool2.get_shape().as_list()
# 计算将矩阵拉直成向量后的长度,这个长度就是矩阵的长宽及深度的乘积,注意这里的pool_shape[0]为一个batch中数据的个数
nodes = pool_shape[1]*pool_shape[2]*pool_shape[3]
# 通过tf.shape函数将第四层的输出变成一个batch的向量
reshaped = tf.reshape(pool2, [pool_shape[0], nodes])
# dropout一般只在全连接层而不是卷积层或者池化层使用
with tf.variable_scope('layer5-fc1'):
fc1_weights = tf.get_variable("weight", [nodes, FC_SIZE],
initializer=tf.truncated_normal_initializer(stddev=0.1))
# 只有全连接层的权重需要加入正则化
if regularizer != None: tf.add_to_collection('losses', regularizer(fc1_weights))
fc1_biases = tf.get_variable("bias", [FC_SIZE], initializer=tf.constant_initializer(0.1))
fc1 = tf.nn.relu(tf.matmul(reshaped, fc1_weights) + fc1_biases)
# 如果train标签为真,则引入dropout函数使输出层一半的神经元失活
if train: fc1 = tf.nn.dropout(fc1, 0.5)
with tf.variable_scope('layer6-fc2'):
fc2_weights = tf.get_variable("weight", [FC_SIZE, NUM_LABELS],
initializer=tf.truncated_normal_initializer(stddev=0.1))
if regularizer != None: tf.add_to_collection('losses', regularizer(fc2_weights))
fc2_biases = tf.get_variable("bias", [NUM_LABELS], initializer=tf.constant_initializer(0.1))
logit = tf.matmul(fc1, fc2_weights) + fc2_biases
return logit
训练基于LeNet的MNIST模型
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
import LeNet5_infernece
import os
import numpy as np
# #### 1. 定义神经网络相关的参数
BATCH_SIZE = 100 # 批处理数量大小
LEARNING_RATE_BASE = 0.01 # 基础学习率
LEARNING_RATE_DECAY = 0.99 # 学习率衰减速率
REGULARIZATION_RATE = 0.0001 # 正则化参数
TRAINING_STEPS = 6000 # 训练周期数
MOVING_AVERAGE_DECAY = 0.99 # 平均滑动步长
# #### 2. 定义训练过程
def train(mnist):
# 定义输出为4维矩阵的placeholder
x = tf.placeholder(tf.float32, [
BATCH_SIZE,
LeNet5_infernece.IMAGE_SIZE,
LeNet5_infernece.IMAGE_SIZE,
LeNet5_infernece.NUM_CHANNELS],
name='x-input')
# y_表示正确的标签
y_ = tf.placeholder(tf.float32, [None, LeNet5_infernece.OUTPUT_NODE], name='y-input')
# 定义L2正则化
regularizer = tf.contrib.layers.l2_regularizer(REGULARIZATION_RATE)
y = LeNet5_infernece.inference(x, False, regularizer) # 表示不使用dropout,但是使用正则化
global_step = tf.Variable(0, trainable=False)
# 定义损失函数、学习率、滑动平均操作以及训练过程。
variable_averages = tf.train.ExponentialMovingAverage(MOVING_AVERAGE_DECAY, global_step)
# 使用平均滑动模型
variables_averages_op = variable_averages.apply(tf.trainable_variables())
# 定以交叉熵函数
cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=y, labels=tf.argmax(y_, 1))
cross_entropy_mean = tf.reduce_mean(cross_entropy)
# 将权重的L2正则化部分加到损失函数中
loss = cross_entropy_mean + tf.add_n(tf.get_collection('losses'))
# 定义递减的学习率
learning_rate = tf.train.exponential_decay(
LEARNING_RATE_BASE,
global_step,
mnist.train.num_examples/BATCH_SIZE, LEARNING_RATE_DECAY,
staircase=True)
train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss, global_step=global_step)
# with tf.control_dependencies([train_step, variables_averages_op]):
# train_op = tf.no_op(name='train')
# 在反向传播梯度下降的过程中更新变量的滑动平均值
train_op = tf.group(train_step, variables_averages_op)
# 初始化TensorFlow持久化类。
saver = tf.train.Saver()
with tf.Session() as sess:
tf.global_variables_initializer().run()
for i in range(TRAINING_STEPS):
xs, ys = mnist.train.next_batch(BATCH_SIZE)
reshaped_xs = np.reshape(xs, (
BATCH_SIZE,
LeNet5_infernece.IMAGE_SIZE,
LeNet5_infernece.IMAGE_SIZE,
LeNet5_infernece.NUM_CHANNELS))
_, loss_value, step = sess.run([train_op, loss, global_step], feed_dict={x: reshaped_xs, y_: ys})
if i%1000 == 0:
print("After %d training step(s), loss on training batch is %g."%(step, loss_value))
# #### 3. 主程序入口
def main(argv=None):
mnist = input_data.read_data_sets("./MNIST_data", one_hot=True)
train(mnist)
if __name__ == '__main__':
main()