lanl-ansi / rosetta-opf

AC-OPF Implementations in Various NLP Modeling Frameworks
Other
52 stars 12 forks source link

Add an Artifacts.toml? #67

Open amontoison opened 4 months ago

amontoison commented 4 months ago

For some research papers with François Pacaud and Sungho Shin, we rely on an Artifacts.toml to easily test our optimization solvers on ACOPF problems. It can be also useful here. The content of Artifacts.toml is:

[PGLib_opf]
git-tree-sha1 = "0e8968a89b6ad43910a8eda4ec30656add35cf91"
lazy = false
    [[PGLib_opf.download]]
    sha256 = "f1421ce22f0a7b9de8a8b2111776b496348220192ad24aace392c3bf608706c2"
    url = "https://github.com/power-grid-lib/pglib-opf/archive/refs/tags/v23.07.tar.gz"

In our test, we can afterward do:

pglib_path = joinpath(artifact"PGLib_opf", "pglib-opf-23.07")

using PowerModels
pglib_opf_case78484_epigrids = ac_power_model(joinpath(pglib_path, "pglib_opf_case78484_epigrids.m"))
odow commented 1 week ago

There's https://github.com/noahrhodes/PGLib.jl

import PGLib
filter(endswith(".m"), readdir(PGLib.PGLib_opf; join = true))

cc @noahrhodes

amontoison commented 1 week ago

I was not aware of PGLib.jl! We should maybe use it as a dependency here? In all cases, a note in the README seems relavant.

noahrhodes commented 1 week ago

I have done very little advertising for it, but it is quite useful for portable benchmarks like this. And as a post-doc now working with LANL, I can garuntee it will be continue to be supported.