| 
 import tensorflow as tf import numpy as np import matplotlib.pyplot as plt  
#使用numpy生成200个随机点,范围从-0.5到0.5均匀分布,增加一个维度得到200行1列的数据(生成二维数据) x_data = np.linspace(-0.5,0.5,200)[:,np.newaxis] #生成随机噪声,形状和x_data相同 noise = np.random.normal(0,0.02,x_data.shape) y_data = np.square(x_data)+noise  
#定义连个placeholder,行不确定,列为1 x = tf.placeholder(tf.float32,[None,1]) y = tf.placeholder(tf.float32,[None,1])  
#定义神经网络中间层 #权值随机数,1行(输入层1个神经元),10列(中间层10个神经元) Weights_L1 = tf.Variable(tf.random_normal([1,10])) #10个偏置值 biases_L1 = tf.Variable(tf.zeros([1,10])) Wx_plus_b_L1 = tf.matmul(x,Weights_L1)+biases_L1 L1 = tf.nn.tanh(Wx_plus_b_L1)  
#定义神经网络输出层 Weights_L2 = tf.Variable(tf.random_normal([10,1])) #1个偏置值 biases_L2 = tf.Variable(tf.zeros([1,1])) Wx_plus_b_L2 = tf.matmul(L1,Weights_L2)+biases_L2 prediction = tf.nn.tanh(Wx_plus_b_L2)  
#二次代价函数 loss = tf.reduce_mean(tf.square(y-prediction)) #梯度下降法 train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)  
with tf.Session() as sess:    #变量初始化    sess.run(tf.global_variables_initializer())    #训练2000次,使用placeholder往x,y 传入x_data,y_data    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()    #散点图    plt.scatter(x_data,y_data)    #红色的实线,宽度为5    plt.plot(x_data,prediction_value,'r-',lw=5)    plt.show()  
 
  |