网络结构：

网络设计：

 

 

 

 代码：

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


tf.compat.v1.disable_eager_execution()
# 1、利用数据，在训练的时候实时提供数据
# mnist手写数字数据在运行时候实时提供给给占位符


tf.compat.v1.app.flags.DEFINE_integer("is_train", 1, "指定是否是训练模型，还是拿数据去预测")
FLAGS = tf.compat.v1.app.flags.FLAGS


def create_weights(shape):
    return tf.Variable(initial_value=tf.compat.v1.random_normal(shape=shape, stddev=0.01))


def create_model(x):
    """
    构建卷积神经网络
    :param x:
    :return:
    """
    # 1）第一个卷积大层
    with tf.compat.v1.variable_scope("conv1"):

        # 卷积层
        # 将x[None, 784]形状进行修改
        input_x = tf.reshape(x, shape=[-1, 28, 28, 1])
        # 定义filter和偏置
        conv1_weights = create_weights(shape=[5, 5, 1, 32])
        conv1_bias = create_weights(shape=[32])
        conv1_x = tf.compat.v1.nn.conv2d(input=input_x, filter=conv1_weights, strides=[1, 1, 1, 1], padding="SAME") + conv1_bias

        # 激活层
        relu1_x = tf.nn.relu(conv1_x)

        # 池化层
        pool1_x = tf.compat.v1.nn.max_pool(value=relu1_x, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding="SAME")

    # 2）第二个卷积大层
    with tf.compat.v1.variable_scope("conv2"):

        # 卷积层
        # 定义filter和偏置
        conv2_weights = create_weights(shape=[5, 5, 32, 64])
        conv2_bias = create_weights(shape=[64])
        conv2_x = tf.compat.v1.nn.conv2d(input=pool1_x, filter=conv2_weights, strides=[1, 1, 1, 1], padding="SAME") + conv2_bias

        # 激活层
        relu2_x = tf.compat.v1.nn.relu(conv2_x)

        # 池化层
        pool2_x = tf.compat.v1.nn.max_pool(value=relu2_x, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding="SAME")

    # 3）全连接层
    with tf.compat.v1.variable_scope("full_connection"):
        # [None, 7, 7, 64]->[None, 7 * 7 * 64]
        # [None, 7 * 7 * 64] * [7 * 7 * 64, 10] = [None, 10]
        x_fc = tf.reshape(pool2_x, shape=[-1, 7 * 7 * 64])
        weights_fc = create_weights(shape=[7 * 7 * 64, 10])
        bias_fc = create_weights(shape=[10])
        y_predict = tf.matmul(x_fc, weights_fc) + bias_fc

    return y_predict


def full_connected_mnist():
    """
    单层全连接神经网络识别手写数字图片
    特征值：[None, 784]
    目标值：one_hot编码 [None, 10]
    :return:
    """
    mnist = input_data.read_data_sets("../mnist_data/", one_hot=True)
    # 1、准备数据
    # x [None, 784] y_true [None. 10]
    with tf.compat.v1.variable_scope("mnist_data"):
        x = tf.compat.v1.placeholder(tf.float32, [None, 784])
        y_true = tf.compat.v1.placeholder(tf.int32, [None, 10])

    y_predict = create_model(x)

    # 3、softmax回归以及交叉熵损失计算
    with tf.compat.v1.variable_scope("softmax_crossentropy"):
        # labels:真实值 [None, 10]  one_hot
        # logits:全脸层的输出[None,10]
        # 返回每个样本的损失组成的列表
        loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y_true, logits=y_predict))

    # 4、梯度下降损失优化
    with tf.compat.v1.variable_scope("optimizer"):
        # 学习率
        train_op = tf.compat.v1.train.AdamOptimizer(0.001).minimize(loss)

    # 5、得出每次训练的准确率（通过真实值和预测值进行位置比较，每个样本都比较）
    with tf.compat.v1.variable_scope("accuracy"):
        equal_list = tf.equal(tf.argmax(y_true, 1), tf.argmax(y_predict, 1))
        accuracy = tf.reduce_mean(tf.cast(equal_list, tf.float32))

    # （2）收集要显示的变量
    # 先收集损失和准确率
    #tf.summary.scalar("losses", loss)
    #tf.summary.scalar("acc", accuracy)

    # 初始化变量op
    init_op = tf.compat.v1.global_variables_initializer()

    # （3）合并所有变量op
    #merged = tf.compat.v1.summary.merge_all()

    # 创建模型保存和加载
    #saver = tf.compat.v1.train.Saver()

    # 开启会话去训练
    with tf.compat.v1.Session() as sess:
        # 初始化变量
        sess.run(init_op)

        # （1）创建一个events文件实例
        #file_writer = tf.compat.v1.summary.FileWriter("../tmp/summary/", graph=sess.graph)

        # 加载模型
        # if os.path.exists("./tmp/modelckpt/checkpoint"):
        #     saver.restore(sess, "./tmp/modelckpt/fc_nn_model")

        if FLAGS.is_train == 1:
            # 循环步数去训练
            for i in range(500):
                # 获取数据，实时提供
                # 每步提供50个样本训练
                mnist_x, mnist_y = mnist.train.next_batch(50)
                # 运行训练op
                sess.run(train_op, feed_dict={x: mnist_x, y_true: mnist_y})
                print("训练第%d步的准确率为：%f, 损失为：%f " % (i+1,
                                     sess.run(accuracy, feed_dict={x: mnist_x, y_true: mnist_y}),
                                     sess.run(loss, feed_dict={x: mnist_x, y_true: mnist_y})
                                     )
                  )

                # 运行合变量op，写入事件文件当中
                #summary = sess.run(merged, feed_dict={x: mnist_x, y_true: mnist_y})
                #file_writer.add_summary(summary, i)
                # if i % 100 == 0:
                #     saver.save(sess, "./tmp/modelckpt/fc_nn_model")

        else:
            # 如果不是训练，我们就去进行预测测试集数据
            for i in range(100):
                # 每次拿一个样本预测
                mnist_x, mnist_y = mnist.test.next_batch(1)
                print("第%d个样本的真实值为：%d, 模型预测结果为：%d" % (
                                                      i+1,
                                                      tf.argmax(sess.run(y_true, feed_dict={x: mnist_x, y_true: mnist_y}), 1).eval(),
                                                      tf.argmax(sess.run(y_predict, feed_dict={x: mnist_x, y_true: mnist_y}), 1).eval()
                                                      )
                                                      )

    return None


if __name__ == "__main__":
    full_connected_mnist()