【Tensorflow】1.一个神经元网络

%conflig IPCompleter.greedy = True    #TAB键自动代码提醒

from tensorflow import keras
import numpy as np
# 构建模型
model = keras.Sequential([keras.layers.Dense(units = 1,input_shape = [1])]) #layers:一层神经元；input_shape：输入值
model.compile(optimizer = 'sgd',loss = 'mean_squared_error') #优化函数 损失函数
# 准备训练数据
xs = np.array([-1.0, 0.0, 1.0, 2.0, 3.0, 4.0], dtype = float)
ys = np.array([-3.0, -1.0, 1.0, 3.0, 5.0, 7.0], dtype = float)
# 训练模型
model.fit(xs,ys,epochs=500)
# 使用模型
print(model.predict([10.0]))

Epoch 1/500 1/1 [==============================] - 0s 290ms/step - loss: 1.1536 Epoch 2/500 1/1 [==============================] - 0s 4ms/step - loss: 1.0379 Epoch 3/500 1/1 [==============================] - 0s 4ms/step - loss: 0.9442 Epoch 4/500 1/1 [==============================] - 0s 3ms/step - loss: 0.8678 Epoch 5/500 1/1 [==============================] - 0s 4ms/step - loss: 0.8052 Epoch 6/500 1/1 [==============================] - 0s 4ms/step - loss: 0.7534 Epoch 7/500 1/1 [==============================] - 0s 4ms/step - loss: 0.7102 Epoch 8/500 1/1 [==============================] - 0s 4ms/step - loss: 0.6738 Epoch 9/500 1/1 [==============================] - 0s 4ms/step - loss: 0.6428 Epoch 10/500 1/1 [==============================] - 0s 3ms/step - loss: 0.6161 Epoch 11/500 1/1 [==============================] - 0s 3ms/step - loss: 0.5928 Epoch 12/500 1/1 [==============================] - 0s 4ms/step - loss: 0.5723 Epoch 13/500

...

1/1 [==============================] - 0s 5ms/step - loss: 2.1943e-05 Epoch 500/500 1/1 [==============================] - 0s 6ms/step - loss: 2.1492e-05 [[18.986473]]

print(model.predict([10.0]))

 [[18.979288]]

model.predict([10.0])

array([[18.979288]], dtype=float32) 

numpy(Numerical Python)： Python 语言的一个扩展程序库，支持大量的维度数组与矩阵运算，此外也针对数组运算提供大量的数学函数库；

keras.layers.Dense()：【Python-Keras】keras.layers.Dense层的解析与使用_Better Bench的博客-CSDN博客_keras.layers.dense

keras.Sequential()：(1条消息) keras系列︱Sequential与Model模型、keras基本结构功能（一）_悟乙己的博客-CSDN博客_sequential

model.compile()：tensorflow中model.compile()用法_yunfeather的博客-CSDN博客_model.compile ；

np.array()：(1条消息) np.array()函数_勤奋的大熊猫的博客-CSDN博客_np.array函数

model.fit()：tensorflow中model.fit()用法_yunfeather的博客-CSDN博客_model.fit函数使用方法；