The Session graph is empty in machine learning pipeline example

[anishsharma]

Hi
I am facing the below error .

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

File "/home/test/dist-keras-021/distkeras/predictors.py", line 23, in __init__ self.model = serialize_keras_model(keras_model) File "/home/test/dist-keras-021/distkeras/utils.py", line 82, in serialize_keras_model print("model.get_weights()=",model.get_weights())

I tried to print the weights but interestingly it is giving exception at the same statement. It does print the weights where we created the model. I am trying an example of yours for machine learning production pipeline with spark streaming.