Sorry, I meant complex as in "very large and complicated expressions" not as in "elements with both real and imaginary components". I'll state a few explicit questions below
How do Theano compilation times scale with problem size? I have computations with hundreds of thousands of nodes. What will this break in Theano? For example are there optimizations that I should exclude when working with FunctionGraphs of this size?
These are only scalar codes so I don't gain much of what Theano has to offer for array computation. Is Theano still a useful tool? One might guess that Theano would introduce overhead that is insignificant for array computations but is significant for scalar computations. Is this the case?
I've provided a link to a pre-existing code. I'm pretty sure it's near optimal and don't expect an automated solution to beat it. Will Theano be able to come close to this code in terms of performance? If not why not?