2.keras-構建基本網絡實現非線性迴歸

構建基本網絡實現非線性迴歸

1.加載顯示數據集

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

# 顯示
plt.scatter(x_data,y_data)
plt.show()

2.構建網絡輸出結果

# 構建順序模型
model = Sequential()
# 在模型中添加一個全鏈接模型
# 機構爲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輸入和輸出都是一維的
# 自定義SGD
sgd = SGD(lr=0.3)
model.compile(optimizer=sgd,
              loss= 'mse')
for step in range(3000):
    # 每次訓練一個batch
    cost = model.train_on_batch(x_data,y_data)
    if step % 500 ==0:
        print('step:',step)
        print('cost',cost)
# 打印權值和偏移項
W,b = model.layers[0].get_weights()
print('W:',W,'b',b)

out:

step: 0
cost 0.066955164
step: 500
cost 0.0051592756
step: 1000
cost 0.019756123
step: 1500
cost 0.0018320761
step: 2000
cost 0.0007798174
step: 2500
cost 0.0005237385
W: [[-0.06731744 0.8597639 0.4614085 0.02440587 -0.04702926 -0.03291976
0.78343517 -0.0447227 1.1036808 1.4795449 ]] b [-0.04047519 0.27002558 -0.06009897 -0.20481145 -0.13842463 -0.27928182
0.21476284 0.28802755 0.44497478 -0.59868914]

3.預測並繪製預測結果

# 進行預測值
y_pred = model.predict(x_data)

# 顯示隨機點
plt.scatter(x_data,y_data)
plt.plot(x_data,y_pred,'r-',lw=3)
plt.show()