News for hackers. Stuff that matters.
Re:L2 Regularization and Batch Norm by phonebucket on Tuesday July 23, 2019 @04:40AM (#20505311)
Jane Street is noted for its use of OCaml, so it's interesting to see that their researchers do indeed use Python (judging from the code in that post, at least).
Re:L2 Regularization and Batch Norm by joshvm on Tuesday July 23, 2019 @05:07AM (#20505402)
They've also used OCaml for deep learning: https://blog.janestreet.com/deep-learning-experiments-in-oca...

Although it seems to be an OCaml binding into TF, rather than a native implementation.

Re:L2 Regularization and Batch Norm by stochastic_monk on Wednesday July 24, 2019 @08:22PM (#20521495)
There’s a similar tch-rs project wrapping libtorch, and in general, deeclaring neural networks is particularly intuitive in functional languages.
Re:L2 Regularization and Batch Norm by xvilka on Tuesday July 23, 2019 @12:38PM (#20508046)
There is a project that aims to bring NN/ML/DL/RL (along with other scientific computations) in the OCaml world - Owl[1]. They also have a list[2] of the potentially interesting ideas for new contributors to take.

[1] http://ocaml.xyz/

[2] http://ocaml.xyz/project/proposal.html

Re:L2 Regularization and Batch Norm by Rainymood on Tuesday July 23, 2019 @05:05AM (#20505396)
Could it be that Python is being used for rapid prototyping but OCaml for production? I've heard of several trading firms with this approach.