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()