https://blog.csdn.net/weixin_44517301/article/details/89605496

https://blog.csdn.net/weixin_44517301/article/details/90176513?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.nonecase&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.nonecase

https://blog.csdn.net/weixin_44517301/article/details/96325200?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.nonecase&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1.nonecase

课堂纪要.py day01_deeplearning.py 深度学习day01.pdf

1 machine_learn VS deeping learn

1.1.1 特征提取

1.1.2 数据量 和 计算性能要求

1.1.3 算法代表

1.2 应用场景

1.3 tensorflow 总体框架

1.3.3 tensorflow install

用Tensorflow实现加法运算演示数据流图（需要开启会话）

import tensorflow  as tf
def tensorflow_demo():
    # Tensorflow实现加法
    a=tf.constant(2)
    b=tf.constant(3)
    c=a+b
    print("Tensorflow加法运算的结果:\n",c)
    #开启会话
    with tf.Session() as sess:
        c_t=sess.run(c)
        print("c_t:\n",c_t)
if __name__ == '__main__':
    tensorflow_demo()

2 Tensorflow 结构分析

2.2 图 与 TensorBoard

2.2.1 什么是图结构

2.2.2 图 相关操作

import tensorflow  as tf
def tensorflow_demo():
    # Tensorflow实现加法
    a=tf.constant(2)
    b=tf.constant(3)
    #c=a+b(不提倡直接使用符合运算)
    c=tf.add(a,b)
    print("Tensorflow加法运算的结果:\n",c)
    # 查看默认图
    # 方法1：调用方法
    default_g = tf.get_default_graph()
    print("default_g:\n", default_g)
    # 方法2:查看属性
    print("a的图属性:\n", a.graph)
    print("c的图属性:\n", c.graph)
    #开启会话
    with tf.Session() as sess:
        c_t=sess.run(c)
        print("c_t:\n",c_t)
        print("sess的图属性:\n",sess.graph)
    return None
if __name__ == '__main__':
    tensorflow_demo()

2.2.3 创建图

import tensorflow  as tf
def tensorflow_demo():
    new_g = tf.get_default_graph()
    with new_g.as_default():
        a_new = tf.constant(20)
        b_new = tf.constant(30)
        c_new = a_new+b_new
        print("c_new:\n",c_new)
    #这时就不能用默认的sesstion了
    #开启new_g的会话
    with tf.Session(graph=new_g) as new_sess:
        c_new_value = new_sess.run(c_new)
        print("c_new_value:\n",c_new_value)
        print("new_sess的图属性:\n",new_sess.graph)
    return None
if __name__ == '__main__':
    tensorflow_demo()

2.2.4 TensorBoard: 可视化

OP

2.3 session

2.3.1 session

2.4 张量

2.4.1 tensor

2.4.2 创建 张量

2.4.3 张量的变换

2.4.4 张量的数学运算

2.5 变量

2.5.1 创建变量

2.5.2 tf.variable 修改变量的命名空间

创建变量、变量的初始化、修改命名空间代码：

import tensorflow as tf
import os
os.environ['TF_CPP_MIN_LOG_LEVEL']='2'
def variable_demo():
    """
    变量的演示
    :return:
    """
    # 创建变量
    with tf.variable_scope("my_scope"):
        a = tf.Variable(initial_value=50)
        b = tf.Variable(initial_value=40)
    with tf.variable_scope("your_scope"):
        c = tf.add(a, b)
    print("a:\n", a)
    print("b:\n", b)
    print("c:\n", c)
    # 初始化变量
    init = tf.global_variables_initializer()
    # 开启会话
    with tf.Session() as sess:
        # 运行初始化
        sess.run(init)
        a_value, b_value, c_value = sess.run([a, b, c])
        print("a_value:\n", a_value)
        print("b_value:\n", b_value)
        print("c_value:\n", c_value)
    return None
if __name__ == '__main__':
    variable_demo()

2.6 高级API

2.6.1 其他基础 API

2.6.2 高级API

2.7 实现线性回归

2.7.1 线性回归原理复习

3 步骤

import tensorflow as tf
def linear_regression():
    """
    自实现一个线性回归
    :return:
    """
    # 1）准备数据
    X = tf.random_normal(shape=[100, 1])
    y_true = tf.matmul(X, [[0.8]]) + 0.7
    # 2）构造模型
    # 定义模型参数 用 变量
    weights = tf.Variable(initial_value=tf.random_normal(shape=[1, 1]))
    bias = tf.Variable(initial_value=tf.random_normal(shape=[1, 1]))
    y_predict = tf.matmul(X, weights) + bias
    # 3）构造损失函数
    error = tf.reduce_mean(tf.square(y_predict - y_true))
    # 4）优化损失
    optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.01).minimize(error)
    # 显式地初始化变量
    init = tf.global_variables_initializer()
    # 开启会话
    with tf.Session() as sess:
        # 初始化变量
        sess.run(init)
        # 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
if __name__ == '__main__':
    linear_regression()

import tensorflow as tf
import os
os.environ['TF_CPP_MIN_LOG_LEVEL']='2'
def linear_regression():
    """
    自实现一个线性回归
    :return:
    """
    with tf.variable_scope("prepare_data"):
        # 1）准备数据
        X = tf.random_normal(shape=[100, 1], name="feature")
        y_true = tf.matmul(X, [[0.8]]) + 0.7
    with tf.variable_scope("create_model"):
        # 2）构造模型
        # 定义模型参数 用 变量
        weights = tf.Variable(initial_value=tf.random_normal(shape=[1, 1]), name="Weights")
        bias = tf.Variable(initial_value=tf.random_normal(shape=[1, 1]), name="Bias")
        y_predict = tf.matmul(X, weights) + bias
    with tf.variable_scope("loss_function"):
        # 3）构造损失函数
        error = tf.reduce_mean(tf.square(y_predict - y_true))
    with tf.variable_scope("optimizer"):
        # 4）优化损失
        optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.01).minimize(error)
    # 2_收集变量
    tf.summary.scalar("error", error)
    tf.summary.histogram("weights", weights)
    tf.summary.histogram("bias", bias)
    # 3_合并变量
    merged = tf.summary.merge_all()
    # 创建Saver对象
    saver = tf.train.Saver()
    # 显式地初始化变量
    init = tf.global_variables_initializer()
    # 开启会话
    with tf.Session() as sess:
        # 初始化变量
        sess.run(init)
        # 1_创建事件文件
        file_writer = tf.summary.FileWriter("./tmp/linear", graph=sess.graph)
        # 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()))
        #
        #     # 运行合并变量操作
        #     summary = sess.run(merged)
        #     # 将每次迭代后的变量写入事件文件
        #     file_writer.add_summary(summary, i)
        #
        #     # 保存模型
        #     if i % 10 ==0:
        #         saver.save(sess, "./tmp/model/my_linear.ckpt")
        # 加载模型
        if os.path.exists("./tmp/model/checkpoint"):
            saver.restore(sess, "./tmp/model/my_linear.ckpt")
        print("训练后模型参数为：权重%f，偏置%f，损失为%f" % (weights.eval(), bias.eval(), error.eval()))
    return None
if __name__ == '__main__':
    linear_regression()

import tensorflow as tf
# 1）定义命令行参数
tf.app.flags.DEFINE_integer("max_step", 100, "训练模型的步数")
tf.app.flags.DEFINE_string("model_dir", "Unknown", "模型保存的路径+模型名字")
# 2）简化变量名
FLAGS = tf.app.flags.FLAGS
def command_demo():
    """
    命令行参数演示
    :return:
    """
    print("max_step:\n", FLAGS.max_step)
    print("model_dir:\n", FLAGS.model_dir)
    return None
if __name__ == '__main__':
    command_demo()

在主函数中直接使用

就可以调用main(argv),功能是显示路径