Tensorfow 笔记（二）

简单神经网络

1.生成随机数据

2.定义placehoder来给训练传递数据

3.定义中间层

4.定义输出层

5.定义方差公式

6.选优化器（梯度下降法。。。）

7.创建会话

8.初始化参数

9.训练 

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

# 使用numpy生成200个随机点
x_data = np.linspace(-0.5, 0.5, 200)[:,np.newaxis]
x_data2 = np.linspace(-1, 1, 200)[:,np.newaxis]
noise = np.random.normal(0,0.02,x_data.shape)

y_data = np.square(x_data) + noise

# 定义两个placeholder
x = tf.placeholder(tf.float32,[None,1])
y = tf.placeholder(tf.float32,[None,1])

# 定义神经网络中间层 10个神经元
Weights_L1 = tf.Variable(tf.random_normal([1,10]))  # 权值
baises_L1 = tf.Variable(tf.zeros([1,10]))  # 偏置值
Wx_plus_b_L1 = tf.matmul(x,Weights_L1) + baises_L1  # y = wx+ b
L1 = tf.nn.tanh(Wx_plus_b_L1)  # 得到中间层的输出

# 定义输出层
Weights_L2 = tf.Variable(tf.random_normal([10,1]))  # 10行一列
baises_L2 = tf.Variable(tf.zeros([1,1]))
Wx_plus_b_L2 = tf.matmul(L1,Weights_L2) + baises_L2 # y = W2*L1 + b
prediction = tf.nn.tanh(Wx_plus_b_L2)

# 二次代价函数
loss = tf.reduce_mean(tf.square(y-prediction))
# 使用梯度下降法训练，最小化loss
train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    for _ in range(2000):
        sess.run(train_step,feed_dict={x:x_data,y:y_data})
    # 获得预测值
    prediction_value = sess.run(prediction,feed_dict={x:x_data})
    # 画图
    plt.figure()
    plt.scatter(x_data,y_data)
    plt.plot(x_data2,prediction_value,'r-',lw = 5)