Tensorflow搭建一个神经网络

一、Tensorlow结构

import tensorflow as tf
import numpy as np

#创建数据
x_data = np.random.rand(100).astype(np.float32)
y_data = x_data*0.1+0.3

#创建一个 tensorlow 结构
weights = tf.Variable(tf.random_uniform([1], -1.0, 1.0))#一维，范围[-1,1]
biases = tf.Variable(tf.zeros([1]))

y = weights*x_data + biases

loss = tf.reduce_mean(tf.square(y - y_data))#均方差函数

#建立优化器，减少误差，提高参数准确度，每次迭代都会优化
optimizer = tf.train.GradientDescentOptimizer(0.5)#学习率为0.5（<1)
train = optimizer.minimize(loss)#最小化损失函数

#初始化不变量
init = tf.global_variables_initializer()

with tf.Session() as sess:
    sess.run(init)
    #train
    for step in range(201):
        sess.run(train)
        if step % 20 == 0:
            print(step, sess.run(weights), sess.run(biases))

二、session的使用

import tensorflow as tf

matrix1 = tf.constant([[3, 3]])
matrix2 = tf.constant([[2], [2]])

product = tf.matmul(matrix1, matrix2)

#method1
sess = tf.Session()
result2 = sess.run(product)
print(result2)

#method2
# with tf.Session() as sess:
#     result2 = sess.run(product)
#     print(result2)

三、Variable的使用

import tensorflow as tf

state = tf.Variable(0, name = 'counter')#变量初始化
# print(state.name)
one = tf.constant(1)
new_value = tf.add(state, one)
#将state用new_value代替
updata = tf.assign(state, new_value)

#变量激活
init = tf.global_variables_initializer()

with tf.Session() as sess:
    sess.run(init)
    for _ in range(3):
        sess.run(updata)
        print(sess.run(state))