I wish I wouldn't have to submit my PhD thesis in 5 weeks, between this and 
Mike's new work on the julia atom client there's never been so much stuff 
that I would happily try to contribute too...
I'll see how the QuantEcon bit works, but otherwise I might just have a go 
at the gradient implementation (although I've never actually used meta 
programming, so while the linear gradient function might seem trivial, 
implementing it won't be trivial to me!)

Reply via email to