Tensorflow: The Session graph is empty. Python

Question

Hello guys I am using Tensorflow 2.0

and in these lines of code:

import tensorflow as tf
hello = tf.constant('Hello World')
sess = tf.compat.v1.Session()
sess.run(hello) <-- Error in this line

RuntimeError: The Session graph is empty. Add operations to the graph before calling run().

Any idea on how to solve it?

Answer 1

g = tf.Graph() 
with g.as_default():   
  # Define operations and tensors in `g`.   
  hello = tf.constant('hello')   
  assert hello.graph is g
    
sess = tf.compat.v1.Session(graph=g) 
sess.run(hello)

b'hello'

Answer 2

Tensorflow core r2.0 have enabled eager execution by default. so, without changing it we just have to change our code as like as below by Launch the graph in a session.

> with tf.compat.v1.Session() as sess:
>     # Building a graph
>     hello = tf.constant("hello")
>     print(sess.run(hello))

According to the Tensorflow doc..

A default Graph is always registered, and accessible by calling tf.compat.v1.get_default_graph

For this basic operations not required to declare a tf.Graph() , you can define a graph which has more computations and dataset you can define a graph and invoke into the session.

Please Refer: For More informations https://www.tensorflow.org/api_docs/python/tf/Graph https://github.com/OlafenwaMoses/ImageAI/issues/400