ndarray-linalg

Linear algebra package for Rust with rust-ndarray.

LAPACKE Backend

Currently three LAPACKE implementations are supported and tested:

• OpenBLAS
  • needs gfortran (or other Fortran compiler)
• Netlib
  • needs cmake and gfortran
• Intel MKL (non-free license, see the linked page)
  • needs curl There are two ways to link LAPACKE backend:

backend features (recommended)

There are three features corresponding to the backend implementations (openblas / netlib / intel-mkl):

[depencdencies]
ndarray = "0.11"
ndarray-linalg = { version = "0.9", features = ["openblas"] }

link backend crate manually

For the sake of linking flexibility, you can provide LAPACKE implementation (as an extern crate) yourself. You should link a LAPACKE implementation to a final crate (like binary executable or dylib) only, not to a Rust library.

[depencdencies]
ndarray = "0.11"
ndarray-linalg = "0.9"
openblas-src = "0.5" # or another backend of your choice

You must add extern crate to your code in this case:

extern crate ndarray;
extern crate ndarray_linalg;
extern crate openblas_src; // or another backend of your choice

For librarian

If you creating a library depending on this crate, we encourage you not to link any backend for flexibility:

[depencdencies]
ndarray = "0.11"
ndarray-linalg = { version = "0.9", default-features = false }

However, if you hope simplicity instead of the flexibility, you can link your favorite backend in the way described above.

Tests and Examples

To run tests or examples for ndarray-linalg, you must specify the desired backend. For example, you can run the tests with the OpenBLAS backend like this:

cargo test --features=openblas

Examples

See examples directory.

Note that to run an example, you must specify the desired backend. For example, you can run the the solve example with the OpenBLAS backend like this:

cargo run --example solve --features=openblas