Shubhamkar Ayare

It looks like it was blatantly copied from `min(max)imum`. It probably existed before your commits. Unfortunately, it looks like I won't have the time and focus to continue working on...

> code coverage [cl-coveralls](https://github.com/fukamachi/cl-coveralls/blob/master/cl-coveralls.asd) is a bit heavy. On the other hand, [sb-cover](https://www.sbcl.org/manual/index.html#sb_002dcover) is incredibly simple to run locally: ```lisp (require :sb-cover) (declaim (optimize sb-cover:store-coverage-data)) (asdf:load-system "numericals" :force t) (asdf:test-system...

I can reproduce this on SBCL 2.4.5 as well for `dense-numericals` but not `numericals`. I can rule out that this unexpected behavior occurs due to the compiler macro of [pelatdot:typep](https://gitlab.com/digikar/peltadot/-/blob/main/src/form-types-and-optim.lisp?ref_type=heads#L71)....

I can also reproduce this on CCL, so this should not have anything to do with SBCL.

The SBCL bug might be that the error location is incorrect. The place that can produce this error will be the second last condition check involving: ```lisp ((AND T (PELTADOT/TYPES-CORE:TYPEP...

Thank you for reporting! The proposed solution seems good for `cl:array`, since these arrays are always contiguous. It fails with discontiguous arrays permitted by dense-arrays and dense-numericals: ```lisp (dense-numericals:concat (list...

The particular polymorph responsible for this case is here: https://github.com/digikar99/numericals/blob/e4d8995459888eb0a4aa05bf9abf5c6f53dabcad/src/basic-math/sum.lisp#L175 Thankfully the pointers are operating all correct and the issue is with the higher level code! [`dense-arrays-plus-lite:reshape` or equivalently `dense-numericals:reshape`](https://github.com/digikar99/dense-arrays/blob/6d9edde102d3bc183a687e3e5af885e7c636cc0a/plus/lite.lisp#L323)...

Ideally for the long run, I want numericals to take the good from other ecosystems. So, if zero-dimension arrays are good in that they help avoid edge cases, I'll be...

> matmul It seems that `matmul` was present in an earlier version but I missed it while refactoring: https://github.com/digikar99/numericals/blob/2024.01.0/src/ranked-functions/matmul.lisp Perhaps there should be also be an additional separate tests system...

I'm understanding this to mean: ``` out = [condition * then] + [(1 - condition) * else] ``` There are SIMD primitives for multiplication, subtraction, and addition. So, I'll expect...