【Tensorflow】tf.Graph()函数

小鱼儿 2024-04-18 14:09 71阅读 0赞

> ### tf.Graph() 函数非常重要，注意提现在两个方面 ###
> 
>  *  **它可以通过tensorboard用图形化界面展示出来流程结构**
>  *  **它可以整合一段代码为一个整体存在于一个图中**

## **一、声明情况大体有三种:** ##

### **1. tensor:通过张量本身直接出graph** ###

# -*- coding: utf-8 -*-  
    import tensorflow as tf
     
    c = tf.constant(4.0)
     
    sess = tf.Session()
    sess.run(tf.global_variables_initializer())
    c_out = sess.run(c)
    print(c_out)
    print(c.graph == tf.get_default_graph())
    print(c.graph)
    print(tf.get_default_graph())

**输出:**

4.0
    True
    <tensorflow.python.framework.ops.Graph object at 0x7f382f9ef110>
    <tensorflow.python.framework.ops.Graph object at 0x7f382f9ef110>

### 2.通过声明一个默认的，然后定义张量内容，在后面可以调用或保存 ###

# -*- coding: utf-8 -*-  
    import tensorflow as tf
     
    g = tf.Graph()
    with g.as_default():
        c = tf.constant(4.0)
     
    sess = tf.Session(graph=g)
    c_out = sess.run(c)
    print(c_out)
    print(g)
    print(tf.get_default_graph())

**输出:**

4.0
    <tensorflow.python.framework.ops.Graph object at 0x7f65f1cb2fd0>
    <tensorflow.python.framework.ops.Graph object at 0x7f65de447c90>

### 3.通过多个声明，在后面通过变量名来分别调用 ###

# -*- coding: utf-8 -*-  
    import tensorflow as tf
     
    g1 = tf.Graph()
    with g1.as_default():
        c1 = tf.constant(4.0)
     
    g2 = tf.Graph()
    with g2.as_default():
        c2 = tf.constant(20.0)
     
    with tf.Session(graph=g1) as sess1:
        print(sess1.run(c1))
    with tf.Session(graph=g2) as sess2:
        print(sess2.run(c2))

**输出:**

4.0
    20.0

## 二、对graph的操作大体有三种: ##

### 1.保存 ###

# -*- coding: utf-8 -*-  
    import tensorflow as tf
     
    g1 = tf.Graph()
    with g1.as_default():
        # 需要加上名称，在读取pb文件的时候，是通过name和下标来取得对应的tensor的
        c1 = tf.constant(4.0, name='c1')
     
    g2 = tf.Graph()
    with g2.as_default():
        c2 = tf.constant(20.0)
     
    with tf.Session(graph=g1) as sess1:
        print(sess1.run(c1))
    with tf.Session(graph=g2) as sess2:
        print(sess2.run(c2))
     
    # g1的图定义，包含pb的path, pb文件名，是否是文本默认False
    tf.train.write_graph(g1.as_graph_def(),'.','graph.pb',False)

**输出:**

4.0
    20.0

###    并且在当前文件夹下面生成graph.pb文件 ###

### 2.从pb文件中调用 ###

# -*- coding: utf-8 -*-  
    import tensorflow as tf
    from tensorflow.python.platform import gfile
     
    #load graph
    with gfile.FastGFile("./graph.pb",'rb') as f:
        graph_def = tf.GraphDef()
        graph_def.ParseFromString(f.read())
        tf.import_graph_def(graph_def, name='')
     
    sess = tf.Session()
    c1_tensor = sess.graph.get_tensor_by_name("c1:0")
    c1 = sess.run(c1_tensor)
    print(c1)

**输出:**

4.0

### 3.穿插调用 ###

# -*- coding: utf-8 -*-  
    import tensorflow as tf
     
    g1 = tf.Graph()
    with g1.as_default():
        # 声明的变量有名称是一个好的习惯，方便以后使用
        c1 = tf.constant(4.0, name="c1")
     
    g2 = tf.Graph()
    with g2.as_default():
        c2 = tf.constant(20.0, name="c2")
     
    with tf.Session(graph=g2) as sess1:
        # 通过名称和下标来得到相应的值
        c1_list = tf.import_graph_def(g1.as_graph_def(), return_elements = ["c1:0"], name = '')
        print(sess1.run(c1_list[0]+c2))

**输出:**

24.0

###  ###

—————————————————————————————————————————  
原文链接：https://blog.csdn.net/zj360202/article/details/78539464