TensorFlow基础：Fetch和Feed

Fetch：一次性取多个op，同时运行多个op

import tensorflow as tf
import os

os.environ["TF_CPP_MIN_LOG_LEVEL"] = "2"  # 忽略tensorflow警告信息

a = tf.constant(3.0)
b = tf.constant(4.0)
c = tf.constant(5.0)

sub = tf.add(a,b)
mul = tf.multiply(a,sub)

with tf.Session() as sess:
    result = sess.run([sub,mul]) # 一次性进行了加和乘的两个op
    print(result)

Feed：

（1）创建占位符

（2）在会话中喂入数据

import tensorflow as tf
import os

os.environ["TF_CPP_MIN_LOG_LEVEL"] = "2"  # 忽略tensorflow警告信息

# 创建占位符
input1 = tf.placeholder(tf.float32)
input2 = tf.placeholder(tf.float32)
output = tf.multiply(input1,input2)

with tf.Session() as sess:
    print(sess.run(output,feed_dict={input1:3,input2:2})) #运行并给前面的占位符传入值，feed以字典的形式传入

结果：

"C:\Program Files\Python36\python.exe" "F:/深度学习/TesnorFlow教程-练数成金/学习/2-fecth and feed.py"
2.0

Process finished with exit code 0