Yeah I've started using Haskell a lot lately, which doesn't really do for loops per se. It was pretty natural to use `map` though because I was plenty used to N

Yeah I've started using Haskell a lot lately, which doesn't really do for loops per se. It was pretty natural to use `map` though because I was plenty used to Numpy style (ok really it's originally APL style but Numpy is where I picked it up). Numba is pretty nifty for things that are unnatural in this style, where an algorithm is better expressed in an index-dependent style, like e.g. a numerical ODE solver. Sometimes you can get around it with tricks like convolution, but not always.