search

ACEsuit / ACE.jl

Parameterisation of Equivariant Properties of Particle Systems
65 stars 15 forks source link

Unexplained usage of dA for multi-species #7

Closed cortner closed 3 years ago

cortner commented 4 years ago

This relates to issue #6 - there is a remaining step which works around that bug, but it should not be required. Specifically the line

   fill!(dA, zero(eltype(dA)))

in oneparticlebasis.jl should not be required since the 1p basis doesn't add into the storage dA but writes into it. So somewhere we are reading lots of zeros. This suggests also that there are some operations that we shouldn't be performing.

cortner commented 3 years ago

seems no longer relevant for v0.10 onwards.