ndarray-linalg

ndarray-linalg copied to clipboard

For the methods in the `lax` crate which call LAPACK, we need to either add bounds checks (i.e. checking that the provided layout will not cause an out-of-bounds read for...

jturner314

Add pinv function (Moore-Penrose Pseudo-inverse)

3

This is my work in progress adding Moore-Penrose Pseudo-inverse of a Matrices #292. I have added all the tests suggested in @jturner314 https://github.com/rust-ndarray/ndarray-linalg/issues/292#issuecomment-849882407 but it still needs more documentation and...

emiddleton

The `SVDDCInto` trait should require `self` to be `DataMut`, but for the `SVDDC` trait we only need `self` to be `Data` since we do not own it.

benkay86

Adding an Adjoint trait for operators

1

I am writing an optimization library for which I would like several of the routine's inputs to be generic LinearOperators. I was happy to find that this trait abstraction was...

cjblocker

factorize_into fails to compile with mutable views

4

``` let mut data = [[2.0f64; 1]; 1]; let mut array = aview_mut2(&mut data); let factorized = array.factorize_into().unwrap(); ``` Fails with ``` note: the following trait bounds were not satisfied:...

v0lat1le

Example failing on mac os

3

I'm having an issue getting the crate to work reliably on OSX. My product requires some simple linear algebra, using the solve functions. My Cargo.toml looks like this ``` [dependencies]...

cgubbin

This adds tests for: - various ways to call `.solveh()`/`.solveh_into()` - 32-bit element types - complex element types - varying array sizes The tests show that the current implementation is...

jturner314

I've created #294 to add tests for these issues.

jturner314

getting no method named `svd` found for reference `&ArrayBase<OwnedRepr<T>, Dim<[usize; 2]>>`

5

Not sure if this is the write place to ask such questions but I couldn't find anywhere else more specific. I am trying to implement the (Moore-Penrose) pseudo-inverse of a...

emiddleton

Improve stride copying in generalize a bit

1

While still not perfect - that should be solved in upstream ndarray, the new version is an improvement since it checks the compatibility before attempting to construct the array. `.into_raw_vec()`...

bluss