import tensorflow as tf

import numpy as np

import keras

from keras.layers import *

from keras.models import Sequential

import matplotlib.pyplot as plt

from keras.optimizers import SGD

x_data = np.linspace(-0.5,0.5,200)

noise = np.random.normal(0,0.02,x_data.shape)

y_data = np.square(x_data) + noise

# 显示

plt.scatter(x_data,y_data)

plt.show()

# 构建顺序模型

model = Sequential()

# 在模型中添加一个全连接模型

# 机构为1-10-1

model.add(Dense(units=10,input_dim=1,activation='tanh'))

model.add(Dense(units=1,activation='tanh')) #units=1,input_dim=1输入和输出都是一维的

# 自定义SGD

sgd = SGD(lr=0.3)

model.compile(optimizer=sgd,

              loss= 'mse')

for step in range(3000):

    # 每次训练一个batch

    cost = model.train_on_batch(x_data,y_data)

    if step % 500 ==0:

        print('step:',step)

        print('cost',cost)

# 打印权值和偏移项

W,b = model.layers[0].get_weights()

print('W:',W,'b',b)

# 进行预测值

y_pred = model.predict(x_data)

# 显示随机点

plt.scatter(x_data,y_data)

plt.plot(x_data,y_pred,'r-',lw=3)

plt.show()