ndarray-linalg
ndarray-linalg copied to clipboard
rust-ndarray
Linear algebra package for rust-ndarray using LAPACK binding
When I want to multiply two matrices that exceed a certain size with the ndarray-linalg blas backend enabled, it returns a matrix filled with zeros. With the native `dot()` from...
`ndarray-linalg` unconditionally enables the `blas` feature of `ndarray`. AFAICT, this feature only changes performance of ndarray matrix multiplication, hence it should not be mandatory to get `ndarray-linalg` running, right ?...
This is to resolve #267. It is equivalent of `eig(A,B)` in Matlab
I am writing an application where I need to calculate eig values. From your documentation, I could see `eig()` but not `eig(&B)`. Is it available?
I noticed that the `calc_eigenvec` parameter is set to true, even if only the eigenvalues are requested (see [here](https://github.com/rust-ndarray/ndarray-linalg/blob/34f7b998eab0276bbd27fb752258f838a4bc1778/ndarray-linalg/src/eigh.rs#L192)). Is there a reason for this? The calculation should probably much...
Document the lobpcg module, fixes #274: * [x] Make `MagnitudeCorrection` public * [x] Improve documentation for `TruncatedEig` * [x] Improve documentation for `TruncatedSvd` * [x] Add Marchensko-Pastur law test
I tried to use `TruncatedSvd` with a generic Float type but it can not compile as it has to use the `MagnitudeCorrection` trait which is not exported from `ndarray-linalg::lobpcg` module....
Solve currently implements `Ax = b` for known NxN matrix `A` and known N-dimensional vector `b`, resulting in a N-dimensional solution vector `x`. A common generalization is `Ax = B`...
BLAS/LAPACK backends has multi-threading. - openblas-src has `openmp` flag to enable OpenMP - intel-mkl-src has `mkl-static-lp64-iomp` flag to link the multi-threaded static library Problems -------------- How to interact with Rust's...
Metadata
Owner
Metadata
Linear algebra package for rust-ndarray using LAPACK binding