Dragon does rock. Amazing how different processor backends and software stacks are all supported.
I walked away from using Clojure years ago after I transitioned from a Clojure loving customer to customers who didn’t use Clojure at all. Whenever I read something written by Dragan I second guess my decision to give up on Clojure.
This is really a perfect case study of why Clojure is the ultimate language for a single person's productivity. Macros and JVM access are a huge productivity increase compared to other functional languages.
Does anyone know any good resources in learning about selection, construction and evaluation of deep learning network architectures to solve various classes of problems?
To many articles seem to be about how to use a particular stack or platform rather than the NN architectures themselves.
You might want to take a look at the books by Timothy Masters. Though i have not looked at his Deep Learning books, his earlier NN books were full of practical advice on the various models/architectures and how to map them to real-world problems.
A meta-reply. I tend to just go look up the papers a tool references as influences and read them. Those generally give you hooks into the other related words to google.
I walked away from using Clojure years ago after I transitioned from a Clojure loving customer to customers who didn’t use Clojure at all. Whenever I read something written by Dragan I second guess my decision to give up on Clojure.