Ask HN: Re-coding algorithms to improve knowledge on deep learning
I've recently got my hands on Deep Learning book by Ian Goodfellow, Yoshua Bengio and Aaron Courville. For some reason I got the urge to re-write the concepts of the book in code, probably c++ and create a small local library.
However, at the moment I am also spending quite a bit of time on enhancing my prototyping and production grade coding skills.
Due to time constrains I'm unable to do both at the same time, so I was wondering which of the options is more preferable.
My goals are to be able to quickly understand the concepts of new research papers and apply it for prototyping purposes. However, I also want to be able to turn that prototype code into production ready code.
P.S. I also thought about taking a month off from work to do this challenge, but I don't know if vacation and topic at hand mixes well. Any experiences?