Question

使用 with tf.Session(graph = g) as sess: 没有给我一个错误但是如果我使用 sess = tf.session(graph = g) 我收到错误消息

g = tf.Graph()

with g.as_default():
    v1 = tf.Variable(1,name = "v1")
    v2 = tf.Variable(2,name = "v2")



with tf.session(graph = g) as sess:
    sess.run(tf.global_variables_initializer())

上面没有给我任何错误

import tensorflow as tf


g = tf.Graph()

with g.as_default():
    v1 = tf.Variable(1,name = "v1")
    v2 = tf.Variable(2,name = "v2")


sess = tf.Session(graph = g)
sess.run(tf.global_variables_initializer())

但这给我一个错误

Traceback (most recent call last):
  File "/Users/wonjunson/gym/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 300, in __init__
    fetch, allow_tensor=True, allow_operation=True))
  File "/Users/wonjunson/gym/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3490, in as_graph_element
    return self._as_graph_element_locked(obj, allow_tensor, allow_operation)
  File "/Users/wonjunson/gym/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3574, in _as_graph_element_locked
    raise ValueError("Operation %s is not an element of this graph." % obj)
ValueError: Operation name: "init"
op: "NoOp"
 is not an element of this graph.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "maintest.py", line 14, in <module>
    sess.run(tf.global_variables_initializer())
  File "/Users/wonjunson/gym/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 929, in run
    run_metadata_ptr)
  File "/Users/wonjunson/gym/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1137, in _run
    self._graph, fetches, feed_dict_tensor, feed_handles=feed_handles)
  File "/Users/wonjunson/gym/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 471, in __init__
    self._fetch_mapper = _FetchMapper.for_fetch(fetches)
  File "/Users/wonjunson/gym/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 271, in for_fetch
    return _ElementFetchMapper(fetches, contraction_fn)
  File "/Users/wonjunson/gym/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 307, in __init__
    'Tensor. (%s)' % (fetch, str(e)))
ValueError: Fetch argument <tf.Operation 'init' type=NoOp> cannot be interpreted as a Tensor. (Operation name: "init"
op: "NoOp"
 is not an element of this graph.)

但是，我看不出它们之间的巨大差异。

我猜这是因为我正在使用global_variables_initializer。由于我只需要在g中初始化变量。但是，变量仍然包含在global_variables中，不是吗？还是变量在图g中是局部的？

Answer 1

我认为您必须已经阅读Related objects reference

大多数TensorFlow程序从数据流图构建阶段开始。在此阶段，您将调用TensorFlow API函数，这些函数构造新的tf.Operation（节点）和tf.Tensor（边缘）对象，并将它们添加到tf.Graph实例中。

我认为，如果您将第二个示例更改为此，则init操作是图形的一部分。

g = tf.Graph()

with g.as_default():
    v1 = tf.Variable(1,name = "v1")
    v2 = tf.Variable(2,name = "v2")
    init = tf.global_variables_initializer() # Part of the graph now


sess = tf.Session(graph = g)
writer = tf.summary.FileWriter("D:/Development_Avecto/TensorFlow/output", sess.graph)
sess.run(init)
writer.close()

您还可以在tensorboard中将其可视化

tensorboard --logdir D:/Development_Avecto/TensorFlow/output

documentation

Answer 2

您还可以使用 as_default() 限制会话和图形上下文以避免错误：

self.sess = tf.Session(graph=self.graph, config=tf_config)
with self.sess.as_default():
    with self.graph.as_default():
        self.sess.run(tf.global_variables_initializer())

在tensorflow中使用会话和仅使用会话有什么区别

2 个答案: