I posted some answers on the GitHub issue:
https://github.com/mikera/core.matrix/issues/228
But in short, you should use one of the optimised implementations if you
care about performance. Clojure vectors are very flexible, but aren't
really designed for numerical performance. vectorz-clj for ex
Which core.matrix implementation are you using? core.matrix is really just
an API, and has a number of implementations. The default is not
particularly fast, and uses vanilla nested Clojure vectors. However, you
can also set up the Vectorz implementation which is typically fast enough
for my ne
