ArrayInterface.jl icon indicating copy to clipboard operation
ArrayInterface.jl copied to clipboard

Designs for new Base array interface primitives, used widely through scientific machine learning (SciML) and other organizations

Results 62 ArrayInterface.jl issues
Sort by recently updated
recently updated
newest added

Successor to #260. Implements basic tools for accessing multidimensional keys and fancy indexing. The goal is to have a common backend for some of the clearly overlapping traits/method that go...

I'm not sure if/when we'll get Base support for `StaticInt` but it would be nice if everybody at least had the tools to support the most of the stride layout...

We have the tooling to in place to solve #332. I think JuliaSIMD uses `stride_rank` too. So I figured it would make sense to just make this a public function...

@rafaqz and @mcabbott, what do you think about having this method defined here to to interfaces together?

I've come across comments that it takes many months or years before a new array type is dependable because it takes time to catch all the corner cases and bugs....

This may already be contained in ArrayInterface somewhere, but I haven't yet found it. It would be great if we could have a trait that indicates the storage order of...

This is based predominantly on the work by @timholy in [this PR](https://github.com/JuliaLang/julia/pull/42029). The plan is to work this into ArrayInterface.jl in another PR, replacing some of #291 and taking advantage...

This PR creates a new type `SubCartesianIndices` for `unsafe_get_collection(::CartesianIndices, ind)` to avoid allocating a new array. And a `SubCartesianIndices` can also be an index like `CartesianIndices`. This PR creates a...

The current implementation of `setindex` is problematic for the following reasons: * It is not compatible with `OffsetArrays`. * It is not compatible with non-numeric types such as `Array{string}`. *...

https://github.com/JuliaArrays/ArrayInterface.jl/pull/266 showed that everything is now fine locally, but CI does not run on the PR branches correctly. Talked with @giordano on this, as I wasn't able to figure it...