Will I replace python+jax with Julia?

I regularly use python+jax for physics simulation tasks, particularly where any machine learning is involved, or where automatic differentiation is (or is likely to become) useful. Else I prefer fortran/C++, depending on exact circumstances. These are typically small/medium-scale projects with 500-10,000 lines of code. Running on clusters is often relevant (and my local cluster has Julia installed!). I don't have much experience with Julia right now. I'm beginning to use it for a substantial project (likely to occupy a month or two of my time), partly as a trial. Will I, at the end of this trial, decide that Julia should replace python+jax for the majority of my python+jax use case? If so, this question resolves YES. Note that I do use python (uncoupled with jax) for many other tasks (easy scripts, etc etc), and I'm not interested here in whether or not Julia will be a good replacement there. Python+jax is not common enough in my field to make "need to play nice with other people" a very significant differentiator. This question may resolve early if the answer becomes blindingly obvious, but I'm going to try not to resolve until I'm very sure, because I have a tendency to oscillate on these matters.

#	Name	Total profit
1		Ṁ99
2		Ṁ58
3		Ṁ30
4		Ṁ16
5		Ṁ15

🏅 Top traders

People are also trading

Related questions