I’ve just released Flux’s first major update since its release, which brings everything up to speed for 0.6 and nails down APIs for things like RNNs and training.
I’ve also reworked the docs based on user feedback, so now may be a good time for another look if things weren’t clear before. Please keep all the feedback and issues coming!
Excitingly, we have a GSoC student, Zhang Shiwei, helping us improve Flux over the summer. We hope to implement optimisers in a generic way that can be reused by other libraries like Knet (among many other things I’m sure).
Enjoy!