Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Package now requires a Julia built with GPL libraries enabled #157

Open
nickrobinson251 opened this issue Jan 10, 2023 · 2 comments
Open

Comments

@nickrobinson251
Copy link

nickrobinson251 commented Jan 10, 2023

...due to the dependency on SparseMatricesCSR.jl (added in #136)


We're running XGBoost.jl in production but have had to fix the XGBoost.jl version to v2.0.2, as the newer versions depend on SparseMatricesCSR.jl, which requires a Julia built with GPL libraries enabled (at Julia build time), causing the following error on our Julia build:

ERROR: LoadError: UndefVarError: UMFPACK not defined Stacktrace:
 [1] getproperty(x::Module, f::Symbol)
   @ Base ./Base.jl:31
 [2] top-level scope
   @ ~/.julia/packages/SparseMatricesCSR/gQcwh/src/SparseMatrixCSR.jl:143
 [3] include(mod::Module, _path::String)
   @ Base ./Base.jl:420
 [4] include(x::String)
   @ SparseMatricesCSR ~/.julia/packages/SparseMatricesCSR/gQcwh/src/SparseMatricesCSR.jl:1
 [5] top-level scope
   @ ~/.julia/packages/SparseMatricesCSR/gQcwh/src/SparseMatricesCSR.jl:17
 [6] include
   @ ./Base.jl:420 [inlined]
 [7] include_package_for_output(pkg::Base.PkgId, input::String, depot_path::Vector{String}, dl_load_path::Vector{String}, load_path::Vector{String}, concrete_deps::Vector{Pair{Base.PkgId, UInt64}}, source::String)
   @ Base ./loading.jl:1554
 [8] top-level scope
   @ stdin:1

I wonder if the SparseMatricesCSR.jl dependency is necessary and e.g. if there might be an alternative way to achieve the functionality added in #136 without taking on this dependency requiring GPL libraries?

@nickrobinson251
Copy link
Author

i also opened gridap/SparseMatricesCSR.jl#21, and with that SparseMatricesCSR branch the XGBoost.jl tests pass locally on a Julia version built with USE_GPL_LIBS=0 (i've no GPU so no GPU tests run)

@ExpandingMan
Copy link
Collaborator

I'm not in favor of re-implementing SparseMatricesCSR in this package, though I'm totally agnostic about which dependency it uses for it as long as it works. The package is used because the DMatrix returns data in this form. I would say that either try to resolve this in SparseMatricesCSR.jl, or if if you'd like to re-implement it we could potentially switch dependencies.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants