Tensorflow works through the creation of computational graphs. I know about tensorflow in the context of machine learning, specifically deep learning. But I know that tensorflow can also be used for wildly different data processing tasks, such as the solving of partial differential equations.
Can the computational Graphs in tensorflow in principle be used to define any arbitrary computational model? (e.g. as an extreme example, could it in principle be used to design the circuitry of a microprocessor?)
