tensorflow1.x的使用记录

tensorflow1.x 使用原则：

1. 先定义图，搭建骨架
2. 再进行图计算

基本模式

import tensorflow as tf
x = tf.constant(12,dtype= 'float32')
sess= tf.Session()
print(sess.run(x))

指定数据类型dtype

import tensorflow as tf
x = tf.constant(12, dtype= 'float32')
with tf.Session() as sess:
    print(sess.run(x))
'''12.0'''

全局初始化，一开始就初始化值

import tensorflow as tf
x = tf.constant(12, dtype= 'float32')
y = tf.Variable(x + 11)
# 初始化
init = tf.global_variables_initializer()
with tf.Session() as sess:
    sess.run(init)
    print(sess.run(y))

import tensorflow as tf
x = tf.constant([14, 23, 40, 30])
y = tf.Variable(x*2 + 100)
model = tf.global_variables_initializer()
with tf.Session() as sess:
    sess.run(model)
    print(sess.run(y))

placeholder占位符

1. 用于一开始不知道具体形状的操作，填充符使用

import tensorflow as tf
x = tf.placeholder("float", None)
y = x * 10 + 500
with tf.Session() as sess:
    placeX = sess.run(y, feed_dict={x: [0, 5, 15, 25]})
    print(placeX)
    '''
[500. 550. 650. 750.]

2. 用于知道部分形状

import tensorflow as tf
x = tf.placeholder("float", [None, 4])
y = x * 10 + 1
with tf.Session() as session:
    dataX = [[12, 2, 0, -2],
              [14, 4, 1, 0],]
    placeX = session.run(y, feed_dict={x: dataX})
    print(placeX)
    '''
[[ 121.   21.    1.  -19.]
 [ 141.   41.   11.    1.]]


'''
import tensorflow as tf
x = tf.placeholder("float", [2, 4])
y = x * 10 + 1
with tf.Session() as session:
    dataX = [[12, 2, 0, -2],
              [14, 4, 1, 0],]
    placeX = session.run(y, feed_dict={x: dataX})
    print(placeX)
'''
[[ 121.   21.    1.  -19.]
 [ 141.   41.   11.    1.]]
'''
import tensorflow as tf
x = tf.placeholder("float", [2, 4])
y = x * 10 + 1
with tf.Session() as session:
    dataX = [[12, 2, 0, -2],
              [14, 4, 1, 0],]
    placeX = session.run(y, feed_dict={x: dataX})
    print(placeX)
'''
[[ 121.   21.    1.  -19.]
 [ 141.   41.   11.    1.]]
'''

用于初始化权重和bias

# Weight and Bias as Variables as they are to be tuned
W = tf.Variable([2], dtype=tf.float32) 
b = tf.Variable([3], dtype=tf.float32)
# Training dataset that will be fed while training
x = tf.placeholder(tf.float32)
# Linear Model
linear_model = W * x + b