JuliaPy / JuliaPy/PyCall.jl

`LinearAlgebra.BLAS.vendor()` has been deprecated

Aperta
#956 4 commenti 0 reazioni 0 assegnatari Vedi su GitHub
Lingua principale
Julia
Stelle
1.5k
Fork
186
Metriche di merge delle PR
Nessuna PR unita negli ultimi 30g

Descrizione

It looks like the `vendor()` function is deprecated in the following line of code:

https://github.com/JuliaPy/PyCall.jl/blob/9a084eb47c7176242893dd18a1782a1114cef02c/src/numpy.jl#L67

```
julia> import PyCall; PyCall.pyimport("jax").numpy.array([1.0])
┌ Warning: `vendor()` is deprecated, use `BLAS.get_config()` and inspect the output instead
│ caller = npyinitialize() at numpy.jl:67
└ @ PyCall ~/.julia/dev/PyCall/src/numpy.jl:67
...
```
I'm on python 3.9.9, numpy 1.21.4, Julia 1.7.1, and [438e738f] PyCall v1.93.0. This is running on NixOS 21.11.
```
julia> versioninfo()
Julia Version 1.7.1
Commit ac5cc99908 (2021-12-22 19:35 UTC)
Platform Info:
OS: Linux (x86_64-pc-linux-gnu)
CPU: Intel(R) Xeon(R) Platinum 8175M CPU @ 2.50GHz
WORD_SIZE: 64
LIBM: libopenlibm
LLVM: libLLVM-12.0.1 (ORCJIT, skylake-avx512)
```
Here's my shell.nix to reproduce the whole thing:
```nix
let
# Last updated: 1/10/22. From status.nixos.org.
pkgs = import (fetchTarball ("https://github.com/NixOS/nixpkgs/archive/8928525bd8b8cdc1235a92a89b72cbbe5bd8a00d.tar.gz")) { };
in
pkgs.mkShell {
buildInputs = with pkgs; [
# See https://github.com/NixOS/nixpkgs/issues/66716. Necessary for julia to
# be able to download packages.
cacert

julia_17-bin

python3
python3Packages.jax
python3Packages.jaxlib
];

# See https://github.com/JuliaPy/PyCall.jl/issues/952#issuecomment-1005694327
PYTHON = "${pkgs.python3}/bin/python";
}
```

Guida per i contributori

Nessuna guida per i contributori indicizzata per questo repository

Direzione di ricerca

Start with src/numpy.jl at line 67, where npyinitialize() triggers the LinearAlgebra.BLAS.vendor() deprecation warning. Reproduce the issue with the reported Julia and PyCall.pyimport("jax").numpy.array([1.0]) command, then verify that the warning is gone while the NumPy/JAX import still works.

Scritto dal modello di indicizzazione a partire dal testo della issue.

Valutazione

Stack tecnologico
julia, numpy, python
Ambito
backend
Tipo di issue
Bug
Difficoltà
2/5
Tempo stimato
1-3 ore
Stato di attività
Ferma
Chiarezza
Abbastanza chiara
Idoneità per principianti
55/100

Ricevi le nuove issue nella tua casella

Un breve riepilogo di issue GitHub adatte ai principianti.