LeNet-5 模型

修改后的LeNet5_infernece

import tensorflow as tf


INPUT_NODE = 784
OUTPUT_NODE = 10


IMAGE_SIZE = 28
NUM_CHANNELS = 1
NUM_LABELS = 10


CONV1_DEEP = 32
CONV1_SIZE = 5


CONV2_DEEP = 64
CONV2_SIZE = 5


FC_SIZE = 512


def inference(input_tensor, train, regularizer):
    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')
        pool_shape = pool2.get_shape().as_list()
        nodes = pool_shape[1] * pool_shape[2] * pool_shape[3]
        reshaped = tf.reshape(pool2, [pool_shape[0], nodes])


    with tf.variable_scope('layer5-fc1'):
        fc1_weights = tf.get_variable("weight", [nodes, FC_SIZE],
                                      initializer=tf.truncated_normal_initializer(stddev=0.1))
        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)
        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

LeNet5_train

# _*_ coding: utf-8 _*_
import os

import tensorflow as tf
import numpy as np
from tensorflow.examples.tutorials.mnist import input_data

# 加载mnist_inference.py中定义的常量和前向传播的函数
import LeNet5_infernece

# 配置神经网络的参数
BATCH_SIZE = 100
LEARNING_RATE_BASE = 0.8
LEARNING_RATE_DECAY = 0.99
REGULARAZTION_RATE = 0.0001
TRAIN_STEPS = 30000
MOVING_AVERAGE_DECAY = 0.99

#MODEL_SAVE_PATH = "./model/"
#MODEL_NAME = "model3.ckpt"

MODEL_SAVE_PATH="MNIST_model/"
MODEL_NAME="mnist_model"

def train(mnist):
    x = tf.placeholder(tf.float32, [BATCH_SIZE, LeNet5_infernece.IMAGE_SIZE,
                                    LeNet5_infernece.IMAGE_SIZE,
                                    LeNet5_infernece.NUM_CHANNELS], name='x-input')
    y_ = tf.placeholder(tf.float32, [None, LeNet5_infernece.OUTPUT_NODE], name='y-input')

    regularizer = tf.contrib.layers.l2_regularizer(REGULARAZTION_RATE)

    y = LeNet5_infernece.inference(x, train, regularizer)

    global_step = tf.Variable(0, trainable=False)

    variable_average = tf.train.ExponentialMovingAverage(MOVING_AVERAGE_DECAY, global_step)
    variable_average_op = variable_average.apply(
        tf.trainable_variables())
    cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=tf.argmax(y_, 1), logits=y)
    cross_entropy_mean = tf.reduce_mean(cross_entropy)

    loss = cross_entropy_mean + tf.add_n(tf.get_collection('losses'))

    learning_rate = tf.train.exponential_decay(LEARNING_RATE_BASE,
                                               global_step=global_step, decay_steps=mnist.train.num_examples / BATCH_SIZE,
                                               decay_rate=LEARNING_RATE_DECAY)
    train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss, global_step=global_step)

    with tf.control_dependencies([train_step, variable_average_op]):
        train_op = tf.no_op(name='train')

    saver = tf.train.Saver()
    with tf.Session() as sess:
        tf.global_variables_initializer().run()

        for i in range(TRAIN_STEPS):
            xs, ys = mnist.train.next_batch(BATCH_SIZE)
            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: xs, y_: ys})
            if i % 1000 == 0:
                print("After %d training steps, loss on training"
                      "batch is %g" % (step, loss_value))
                saver.save(sess, os.path.join(MODEL_SAVE_PATH, MODEL_NAME), global_step=global_step)
                

if __name__ == '__main__':        
    mnist = input_data.read_data_sets('MNIST_data/', one_hot=True)  
    train(mnist)  

结果：

Extracting MNIST_data/train-images-idx3-ubyte.gz
Extracting MNIST_data/train-labels-idx1-ubyte.gz
Extracting MNIST_data/t10k-images-idx3-ubyte.gz
Extracting MNIST_data/t10k-labels-idx1-ubyte.gz
After 1 training steps, loss on trainingbatch is 6.74421
After 11 training steps, loss on trainingbatch is 18.0097
After 21 training steps, loss on trainingbatch is 17.9667
After 31 training steps, loss on trainingbatch is 17.9231
After 41 training steps, loss on trainingbatch is 17.9186