深度学习05

构造线性回归模型

代码

def linear_regression():
    """
    自实现一个线性回归
    :return:
    """
    # 1）准备数据
    x = tf.compat.v1.random_normal(shape=[100, 1])
    y_true = tf.matmul(x, [[0.8]]) + 0.7
    print("前：",x)
    # 2）构造模型
    # 定义模型参数 用 变量
    weights = tf.Variable(initial_value=tf.compat.v1.random_normal(shape=[1, 1]))
    bias = tf.Variable(initial_value=tf.compat.v1.random_normal(shape=[1, 1]))
    y_predict = tf.matmul(x, weights) + bias
    # 3）构造损失函数 均方误差
    error = tf.reduce_mean(tf.square(y_predict - y_true))
    # 4）优化损失
    optimizer = tf.compat.v1.train.GradientDescentOptimizer(learning_rate=0.01).minimize(error)
    # 显式地初始化变量
    init = tf.compat.v1.global_variables_initializer()
    # 开启会话
    with tf.compat.v1.Session() as sess:
        # 初始化变量
        sess.run(init)
        print("后：",sess.run(y_true))
        # 查看初始化模型参数之后的值
        print("训练前模型参数为：权重%f，偏置%f，损失为%f" % (weights.eval(), bias.eval(), error.eval()))
        #开始训练
        for i in range(100):
            sess.run(optimizer)
            print("第%d次训练后模型参数为：权重%f，偏置%f，损失为%f" % (i+1, weights.eval(), bias.eval(), error.eval()))

    return None

构造逻辑回归模型

代码

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


def full_connection():
    tf.compat.v1.disable_eager_execution()
    # 1）准备数据
    mnist = input_data.read_data_sets("../mnist_data", one_hot=True)
    # 用占位符定义真实数据
    x = tf.compat.v1.placeholder(dtype=tf.float32, shape=[None, 784])
    y_true = tf.compat.v1.placeholder(dtype=tf.float32, shape=[None, 10])

    # 2）构造模型 - 全连接
    # [None, 784] * W[784, 10] + Bias = [None, 10]
    weights = tf.Variable(initial_value=tf.compat.v1.random_normal(shape=[784, 10], stddev=0.01))
    bias = tf.Variable(initial_value=tf.compat.v1.random_normal(shape=[10], stddev=0.1))
    #y_predict = tf.matmul(x, weights) + bias

    y_predict = tf.compat.v1.nn.softmax(tf.matmul(x, weights) + bias)
    loss=tf.reduce_mean(-tf.reduce_sum(y_true*tf.compat.v1.log(y_predict),1))
    # 3）构造损失函数
    #loss_list = tf.nn.softmax_cross_entropy_with_logits(logits=y_predict, labels=y_true)
    #loss = tf.reduce_mean(loss_list)

    # 4）优化损失 梯度下降
    optimizer = tf.compat.v1.train.GradientDescentOptimizer(learning_rate=0.01).minimize(loss)
    #optimizer = tf.compat.v1.train.AdamOptimizer(learning_rate=0.01).minimize(loss)

    # 5）增加准确率计算 argmax(y_true, axis=1)求矩阵行最大值的索引  cast()类型转换
    bool_list = tf.equal(tf.argmax(y_true, axis=1), tf.argmax(y_predict, axis=1))
    accuracy = tf.reduce_mean(tf.cast(bool_list, tf.float32))

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

    # 开启会话
    with tf.compat.v1.Session() as sess:

        # 初始化变量
        sess.run(init)

        # 开始训练
        for i in range(100):
            # 获取真实值
            image, label = mnist.train.next_batch(500)
            # print(image.shape)
            # print(label.shape)
            _, loss_value, accuracy_value = sess.run([optimizer, loss, accuracy], feed_dict={x: image, y_true: label})
            tsst_acc=sess.run(accuracy,feed_dict={x:mnist.test.images,y_true:mnist.test.labels})
            print("第%d次的损失为%f，准确率为%f, 测试集准确率%f" % (i+1, loss_value, accuracy_value,tsst_acc))


    return None

if __name__ == "__main__":
    full_connection()

神经网络模型

代码

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

tf.compat.v1.disable_eager_execution()
mnist = input_data.read_data_sets("./mnist_data", one_hot=True)

# 用占位符定义真实数据
x = tf.compat.v1.placeholder(dtype=tf.float32, shape=[None, 784])
y = tf.compat.v1.placeholder(dtype=tf.float32, shape=[None, 10])

weights={
    'w1':tf.Variable(initial_value=tf.compat.v1.random_normal(shape=[784, 256], stddev=0.01)),
    'w2':tf.Variable(initial_value=tf.compat.v1.random_normal(shape=[256, 128], stddev=0.01)),
    'out':tf.Variable(initial_value=tf.compat.v1.random_normal(shape=[128, 10], stddev=0.01)),
}
biases={
    'b1':tf.Variable(initial_value=tf.compat.v1.random_normal(shape=[256])),
    'b2':tf.Variable(initial_value=tf.compat.v1.random_normal(shape=[128])),
    'out':tf.Variable(initial_value=tf.compat.v1.random_normal(shape=[10])),
}

def multilayer_perceptron(_x,_weights,_biases) :
    layer_1 = tf.compat.v1.nn.sigmoid(tf.add(tf.matmul(_x,_weights['w1']),_biases['b1']))
    layer_2 = tf.compat.v1.nn.sigmoid(tf.add(tf.matmul(layer_1,_weights['w2']),_biases['b2']))
    return (tf.matmul(layer_2,_weights['out']) +_biases['out'])

pred=multilayer_perceptron(x,weights,biases)

loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=pred, labels=y))
# optimizer = tf.compat.v1.train.GradientDescentOptimizer (learning_rate=0.01).minimize(loss)
optimizer = tf.compat.v1.train.AdamOptimizer(learning_rate=0.01).minimize(loss)
bool_list = tf.equal(tf.argmax(pred,1), tf.argmax(y,1))
accuracy = tf.reduce_mean(tf.cast(bool_list,tf.float32))

init = tf.compat.v1.global_variables_initializer()

with tf.compat.v1.Session() as sess:
    # 初始化变量
    sess.run(init)

    # 开始训练
    for i in range(100):
        # 获取真实值
        image, label = mnist.train.next_batch(500)
        # print(image.shape)
        # print(label.shape)
        _, loss_value, accuracy_value = sess.run([optimizer, loss, accuracy], feed_dict={x: image, y: label})
        tsst_acc = sess.run(accuracy, feed_dict={x: mnist.test.images, y: mnist.test.labels})
        print("第%d次的损失为%f，准确率为%f, 测试集准确率%f" % (i + 1, loss_value, accuracy_value, tsst_acc))