tensorflow基本用法示例

关于tensorflow的基本用法，教程详见：
TensorFlow官方文档中文版——基本使用，
和 10 minutes Practical TensorFlow Tutorial for quick learners。
（第二个教程相对更好一点）
由于版本更新，原教程的一些代码已不能使用，下面是修改之后的tensorflow基本用法示例。

一.TensorFlow官方文档中文版——基本使用

1. add_operation

import tensorflow as tf

hello=tf.constant('hello, TensorFlow!')

sess=tf.Session()

print(sess.run(hello))

b'hello, TensorFlow!'

a=tf.constant(10)
b=tf.constant(32)

print(sess.run(a+b))

2. counter

import tensorflow as tf

#创建一个变量，初始化为标量0
state=tf.Variable(0, name='counter')

#创建一个op，其作用是使state增加1
one=tf.constant(1)

new_value=tf.add(state, one)
update=tf.assign(state, new_value)

#启动图后，变量必须先经过‘初始化’(init)op初始化。
#首先必须增加一个‘初始化’op到图中
init_op=tf.global_variables_initializer()

#启动图，运行op
with tf.Session() as sess:
#运行‘init’op
    sess.run(init_op)
#打印‘state’的初始值
    print(sess.run(state))
#运行op,更新‘state’,并打印‘state’
for _ in range(3):
        sess.run(update)
        print(sess.run(state))

3. matrix_multi

import tensorflow as tf

创建一个常量op，产生一个1*2矩阵，这个op被作为一个节点加到默认图中。构造器的返回值代表该常量op的返回值

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

创建另外一个常量op，产生一个2*1矩阵

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

创建一个矩阵乘法matmul op，把’matrix1’和’matrix2’作为输入。返回值’product’代表矩阵乘法的结果。

product = tf.matmul(matrix1, matrix2)

构造阶段完成后，开始启动图，启动图的第一步是创建一个Session对象。Session对象在使用完后需要关闭以释放资源，除了显式调用close外，也可以使用’with’代码自动完成关闭动作.

with tf.Session() as sess:
    result=sess.run([product])
    print(result)

[array([[12]], dtype=int32)]

4. fetch_feed

import tensorflow as tf

input1=tf.constant(3.0)
input2=tf.constant(2.0)
input3=tf.constant(5.0)
intermed=tf.add(input2, input3)
mul=tf.multiply(input1, intermed)

with tf.Session() as sess:
    result=sess.run([mul, intermed])
    print(result)

[21.0, 7.0]

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:[7.], input2:[2.]}))

[array([ 14.], dtype=float32)]

二. 10 minutes Practical TensorFlow Tutorial for quick learners

Basics of TensorFlow

import tensorflow as tf

#you can access this graph by:
graph=tf.get_default_graph()

#you can get the list of all operations by typing this:
#graph.get_operations()
for op in graph.get_operations():
    print(op.name)

Constants

a=tf.constant(1.0)
a

<tf.Tensor 'Const:0' shape=() dtype=float32>

print(a)

Tensor("Const:0", shape=(), dtype=float32)

with tf.Session() as sess:
    print(sess.run(a))

1.0

Variables

b=tf.Variable(2.0, name="test_var")
b

<tensorflow.python.ops.variables.Variable at 0x7fb8b90ce4e0>

init_op=tf.global_variables_initializer()
with tf.Session() as sess:
    sess.run(init_op)
    print(sess.run(b))

2.0

graph=tf.get_default_graph()
for op in graph.get_operations():
    print(op.name)

Const
test_var/initial_value
test_var
test_var/Assign
test_var/read
init

Placeholders

a=tf.placeholder("float")
b=tf.placeholder("float")
y=tf.multiply(a, b)

feed_dict={a:2, b:3}
with tf.Session() as sess:
    print(sess.run(y, feed_dict))

6.0

Tensorflow tutorial with linear regression

#create a random normal distribution:
w=tf.Variable(tf.random_normal([784, 10], stddev=0.01))

#calculates the mean of an array:
b=tf.Variable([10, 20, 30, 40, 50, 60], name='t')
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    print(sess.run(tf.reduce_mean(b)))

#ArgMax:very seimilar to python argmax(numpy.argmax)
a=[[0.1, 0.2, 0.3],
  [20, 2, 3]]
b=tf.Variable(a, name='b')
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    print(sess.run(tf.argmax(b, 1)))

[2 0]

#(a)Creating training data
import tensorflow as tf
import numpy as np

trainX=np.linspace(-1, 1, 101)
trainY=3*trainX+np.random.randn(*trainX.shape)*0.33

#(b)Placeholders
X=tf.placeholder("float")
Y=tf.placeholder("float")

#(c)Modeling
w=tf.Variable(0.0, name="weights")
y_model = tf.multiply(X, w)

cost=tf.pow(Y-y_model, 2)
train_op=tf.train.GradientDescentOptimizer(0.01).minimize(cost)

#(d)Training
init=tf.global_variables_initializer()
with tf.Session() as sess:
    sess.run(init)
for i in range(100):
for (x, y) in zip(trainX, trainY):
            sess.run(train_op, feed_dict={X:x, Y:y})
    print(sess.run(w))

2.96799

#(e)Exercise
with tf.Session() as sess:
    sess.run(init)
    print(sess.run(w))

0.0

秒客网