I have an idea about name scope that naming a group of OPS to view in graph. But I am loosing my mind at variable scope. I can't get it right to execute my code in multiple GPU.
Variable scope allows you to create new variables and to share already created ones while providing checks to not create or share by accident. For details, see the https://www.tensorflow.org/api_docs/python/tf/variable_scope
i think the best way to understand it is through tensorgraph debugger