Christoph Ortner

I've run into this as well. How do I need to restrict the versions of Loopvectorisation and Stridearrays to avoid this? Or at least I think my problem is the...

fantastic, thank you!

JuLIP is quite old and possibly not flexible enough to easily adapt to abstract number types. E.g. I've long wanted to allow dual and hyer-dual numbers. Those would give you...

that's the goal

If you look at EmpiricalPotentials now, the SW potential can already accept general number types as inputs.

We could remove the cutoff check by multiplying with a characteristic function that could then be overloaded for your number type. In principle I'm open to this, but I suspect...

> Is ChemistryFeaturization going to have a more central role, like a one-stop shop of sorts, for every type of chemistry/molecular science-related featurization? I would like that. I'm having a...

Im open for anything really since at our end we have to write wrappers anyhow to make this compatible with how we organize computations.

My only gripe with what you suggest is that the type of the calculator should determine whether it's spherical or solids - hence just "compute". But as a general rule...

I have a prototype implementation, which - on small basis sizes - is about a factor 2-3 faster than my previous one based on spherical coords (for large bases they...