neanderthal
neanderthal copied to clipboard
uncomplicate
Fast Clojure Matrix Library
It setups MKL, CUDA and openCL based on Ubuntu 20.04 Dockerimage. Maybe worth to keep and mention as an additional way of setting up of neanthertal. I went for CUDA...
Intel changed MKL to be part of the oneAPI in the 2021 version. With that change the dll files within MKL changed somewhat. The most notable change is that `mkl_rt.dll`...
I don't know if this is just my setup, or an actual bug in the printing of these vectors. It is a visual/printing issue only I think. Experienced issue: ```...
Would be nice to have "native" support on next-gen macs. Looks like the [Accelerate](https://developer.apple.com/documentation/accelerate/veclib) library provides BLAS and LAPACK
the documentation for `ls` could maybe be clarified a bit. if you have `ax=b` `a` is 3-by-2 `b` is 3-by-1 then `x` should be 2-by-1 `ls!` as well as `ls`...
This also happens if you run a jar on the command line: ```clojure user> (require '[tech.v3.dataset.neanderthal]) 08:32:27.717 [nREPL-session-e83e58b7-6a87-49ee-91b6-5bb82ed7210e] DEBUG org.scijava.nativelib.BaseJniExtractor - mappedLib is libneanderthal-mkl-0.33.0.so 08:32:27.723 [nREPL-session-e83e58b7-6a87-49ee-91b6-5bb82ed7210e] DEBUG org.scijava.nativelib.BaseJniExtractor - URL...
Links in docstrings to the intel documentation (e.g. to https://software.intel.com/en-us/node/468426 in uncomplicate.neanderthal.core) are no longer available. The internet archive saved it, though: https://web.archive.org/web/20181005053902/https://software.intel.com/en-us/node/468426
From the documentation, it seems `org` returns Q in the reduced QR decomposition (e.g. `(org (qrf M))` ). Whereas, MKL's `?orgqr` allows the call to specify the number of columns...
As I understand it, `es!` gives the Schur decomposition of a matrix by calling MKL [`gees`](https://software.intel.com/content/www/us/en/develop/documentation/mkl-developer-reference-c/top/lapack-routines/lapack-least-squares-and-eigenvalue-problem-routines/lapack-least-squares-and-eigenvalue-problem-driver-routines/nonsymmetric-eigenvalue-problems-lapack-driver-routines/gees.html). Unfortunately there is no way to get sorted real Schur form, which is sometimes...
[Please support my work on Patreon](https://patreon.com/draganrocks). I'll intive you to a dedicated [Discord](https://discordapp.com) discussion server. Can't afford to donate? Ask for a free invite.
