import tensorflow as tf
import numpy as np
 
x = tf.Variable(0)
init = tf.initialize_all_variables()
sess = tf.InteractiveSession()
sess.run(init)
 
print(x.eval())
 
x.assign(1)
print(x.eval())

import tensorflow as tf
x = tf.Variable(0)
y = tf.assign(x, 1)
with tf.Session() as sess:
 sess.run(tf.global_variables_initializer())
 print sess.run(x)
 print sess.run(y)
 print sess.run(x)

In [212]: w = tf.Variable(12)
In [213]: w_new = w.assign(34)
 
In [214]: with tf.Session() as sess:
  ...:  sess.run(w_new)
  ...:  print(w_new.eval())
 
# output
34 

import tensorflow as tf
x = tf.Variable(0)
sess = tf.Session()
sess.run(tf.global_variables_initializer())
print(sess.run(x)) # Prints 0.
x.load(1, sess)
print(sess.run(x)) # Prints 1.

import numpy as np #這是Python的一種開源的數(shù)值計算擴展，非常強大
import tensorflow as tf #導(dǎo)入tensorflow 

##構(gòu)造數(shù)據(jù)##
x_data=np.random.rand(100).astype(np.float32) #隨機生成100個類型為float32的值
y_data=x_data*0.1+0.3 #定義方程式y(tǒng)=x_data*A+B
##-------##

##建立TensorFlow神經(jīng)計算結(jié)構(gòu)##
weight=tf.Variable(tf.random_uniform([1],-1.0,1.0)) 
biases=tf.Variable(tf.zeros([1]))  
y=weight*x_data+biases

w1=weight*2

loss=tf.reduce_mean(tf.square(y-y_data)) #判斷與正確值的差距
optimizer=tf.train.GradientDescentOptimizer(0.5) #根據(jù)差距進行反向傳播修正參數(shù)
train=optimizer.minimize(loss) #建立訓(xùn)練器

init=tf.global_variables_initializer() #初始化TensorFlow訓(xùn)練結(jié)構(gòu)
#sess=tf.Session() #建立TensorFlow訓(xùn)練會話
sess = tf.InteractiveSession() 
sess.run(init)  #將訓(xùn)練結(jié)構(gòu)裝載到會話中
print('weight',weight.eval())
for step in range(400): #循環(huán)訓(xùn)練400次
  sess.run(train) #使用訓(xùn)練器根據(jù)訓(xùn)練結(jié)構(gòu)進行訓(xùn)練
  if step%20==0: #每20次打印一次訓(xùn)練結(jié)果
  print(step,sess.run(weight),sess.run(biases)) #訓(xùn)練次數(shù)，A值，B值
  
print(sess.run(loss))  
print('weight new',weight.eval())


#wop=weight.assign([3])
#wop.eval()
weight.load([1],sess)
print('w1',w1.eval())