【tensorflow】1.添加层及搭建神经网络

参考——莫烦python

包括：

1.变量定义

2.session控制

3.占位符

4.添加层

5.搭建神经网络

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time    : 18/2/27 下午5:11
# @Author  : cicada@hole
# @File    : tf.py
# @Desc    : tensorflow测试文件
# @Link    :

import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt

def func1():
    # d
    x_data = np.random.rand(100).astype(np.float32)
    y_data = x_data * 0.1 + 0.3

    # m
    Weights = tf.Variable(tf.random_uniform([1], -1.0, 1.0))
    biases = tf.Variable(tf.zeros([1]))

    y = Weights * x_data + biases

    loss = tf.reduce_mean(tf.square(y - y_data))

    optimizer = tf.train.GradientDescentOptimizer(0.5)
    train = optimizer.minimize(loss)

    init = tf.global_variables_initializer()
    session = tf.Session()
    session.run(init)

    for step in range(201):
        session.run(train)
        if step % 20 == 0:
            print(step, session.run(Weights), session.run(biases))

def sessionControl():
   # 创建两个矩阵
    matrix1 = tf.constant([[3,3],[2,3]])
    matrix2 = tf.constant([[2],[2]])
    product = tf.matmul(matrix1, matrix2)

   #用session来激活product并得到计算结果
    sess = tf.Session()

    # method1
    result = sess.run(product)
    print(result)
    sess.close()

    # method 2
    with tf.Session() as sess:
        result2 = sess.run(product)
        print(result2)

def tfVarDefine():
    # 定义变量 语法： state = tf.Variable()
    state = tf.Variable(0, name='counter')

    # 定义常量 one
    one = tf.constant(1)

    # 定义加法步骤（此步并没有直接计算）
    new_value = tf.add(state, one)

    # 将state 更新为 new_value
    update = tf.assign(state, new_value)

    # 定义的变量需要激活
    init = tf.global_variables_initializer()

    # 使用session
    with tf.Session() as sess:
        sess.run(init)
        for _ in range(3):
            sess.run(update)
            print('---------state ',sess.run(state)) # 需要把sess的指针指向state 再print

#-------占位符placeholder----
def placeHolderTest():
    input1 = tf.placeholder(tf.float32)  #大部分形式f32
    input2 = tf.placeholder(tf.float32)

    output = tf.multiply(input1, input2)  #乘法运算


    with tf.Session() as sess:
        print(sess.run(output, feed_dict={input1:[7.],input2:[2.]}))
        # feed_dict和placeholder进行绑定，能传值

#----激励函数-----如sigmod等
def activationFunctionTest():
    pass

#----添加层 ----add_layer()----
def add_layer(inputs, in_size, out_size, activation_function=None):
    Weights = tf.Variable(tf.random_normal([in_size, out_size])) #in_size行out_size列的随机矩阵
    biases = tf.Variable(tf.zeros([1, out_size]) + 0.1) #biases推荐值不为零
    Wx_plus_b = tf.matmul(inputs, Weights) + biases #神经网络未激活的值
    #activation_function = None时，输出Wx_plus_b,不为None时，输出激活后的Wx_plus_b
    if activation_function is None:
        outputs = Wx_plus_b
    else:
        outputs = activation_function(Wx_plus_b)
    return outputs

#----搭建网络-----
def buildNN():
    #导入数据
    x_data = np.linspace(-1, 1, 300, dtype=np.float32)[:, np.newaxis]
    noise = np.random.normal(0, 0.05, x_data.shape).astype(np.float32)
    y_data = np.square(x_data) - 0.5 + noise

    xs = tf.placeholder(tf.float32, [None, 1])
    ys = tf.placeholder(tf.float32, [None, 1])

    #输入层1个、隐藏层10个、输出层1个的神经网络
    l1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu)#输入层只有一个属性

    #输入就是隐藏层的输出——l1，输入有10层（隐藏层的输出层），输出有1层
    prediction = add_layer(l1, 10, 1, activation_function=None)

    loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction), reduction_indices=[1]))#平均误差
    # reduction_indices表示函数的处理维度

    # 让机器学习提升准确率 以0.1的效率来最小化误差loss
    train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

    init = tf.global_variables_initializer()
    sess = tf.Session()
    sess.run(init)
    fig = plt.figure()
    ax = fig.add_subplot(1, 1, 1)
    ax.scatter(x_data, y_data)
    plt.ion()  # 连续显示
    plt.show()

    for i in range(1000):
        sess.run(train_step, feed_dict={xs: x_data, ys: y_data})
        if i % 50 == 0:
            # to visualize the result and improvement
            try:
                ax.lines.remove(lines[0])
            except Exception:
                pass
            prediction_value = sess.run(prediction, feed_dict={xs: x_data})
            # plot the prediction
            lines = ax.plot(x_data, prediction_value, 'r-', lw=5) #显示预测的数据
            plt.pause(0.1)


if __name__ == '__main__':
    # sessionControl()
    # tfVarDefine()
    # placeHolderTest()
    buildNN()