Clojure data science vs targeting julia

I apologize in advance if this is a naive question, but I have seen some discussion about Clojure data science and wondered if it would be more advantageous to target another language like Julia. Essentially writing Julia in Clojure. Is that a possibility if someone was so inclined, or is it a bad idea? I thought of Julia because it is performant, has a community dedicated to data science, it has macros, and practically bulit on scheme.

Thanks,
Ben

.

1 Like

I think a Clojure dialect for Julia would probably be more work than just improving Clojure’s data-science libraries and tools.

Not that it wouldn’t be cool!

I also saw this recently: https://github.com/cnuernber/libpython-clj

Python bindings from Clojure.

1 Like

There are thoughts to target Julia in the future in a way similar to the Python bindings

Interesting ideas.

Though it is not exactly Clojure hosted in Julia, LispSyntax.jl is worth checking (I never used it seriously, just played a little).

It is not a new project, but seems to be getting some attention recently.

Hello, @Benjamin_Neil!

Welcome to ClojureVerse. Please feel free to add a sentence or two about yourself in the Introduce yourself thread if you feel like it!

To me, it sounds like a good idea to provide language bindings similar to the ones @didibus linked to.

I’d also like to link to a few cool data science libraries written specifically for Clojure:

  • TVM enables writing high-performance numeric code, all in Clojure. Target the CPU or the GPU.
  • Clojure MXNET provides Clojure bindings for the language-agnostic Apache MXNET deep learning library.
  • Oz is a visualization package for Clojure that works really well with REPL-driven development.

Teodor

1 Like