3 ms·
Interesting idea to use an ML for scientific programming, but I don't see any practical reasons not to use Julia or Python. I'd rather take advantage of everyt
by dunefox 6y ago
Interesting idea to use an ML for scientific programming, but I don't see any practical reasons not to use Julia or Python. I'd rather take advantage of everything Julia already offers (+ Python with PyCall.jl) than wait for the same support in a language not widely used in the first place.
- mseri 6y agoYou can use PyML to call python from OCaml in the same way, and it works fine to pass an owl ndarray to numpy. As a user of both, I think they have different treadoffs. I tend to use OCaml when I am playing around with the code because I find it infinitely easier to refactor (and to figure out what I was doing if I leave the code rotten for too long)