Great news for R users who largely have been left behind by the deep learning community.<p>The main downside to this is that we will now start seeing all sorts of experimental models from R users who know a lot more about statistics than about software engineering and therefore often write code that is not very elegant nor easy to understand.[1]<p>[1] Don't take it from me. Quoting Hadley Wickham: "Much of the R code you’ll see in the wild is written in haste to solve a pressing problem. As a result, code is not very elegant, fast, or easy to understand. Most users do not revise their code to address these shortcomings. Compared to other programming languages, the R community tends to be more focussed on results instead of processes. Knowledge of software engineering best practices is patchy." <a href="http://adv-r.had.co.nz/Introduction.html" rel="nofollow">http://adv-r.had.co.nz/Introduction.html</a>
This is a big leap for Machine Learning in R. R keeps moving forward in so many ways. We now have native dplyr into Spark and now a TensorFlow library from RStudio.