Sobhan Mohammadpour

> I hadn't seen the rest of the API here i don't think it makes sense to bring that in SparseArrays. the parallel `\` is nice but i'm not sure...

@dkarrasch The links were very helpful. I think that this works as intended now. ```julia julia> A = Symmetric([randn(20, 20) for _ in 1:20, _ in 1:20]); julia> sum(sum(A[i, j]...

> I thought the main issue here was the significant performance gap. I don't think that this is an issue look at the plots in https://sobhanmp.github.io/SparseExtra.jl/iternz/ The loss of performance...

@rayegun (ping)

but is that a good idea? i think batching it is going to be much faster due to hitting GC/data shift.

It's probably a bit better but what you are describing is not a good idea in general. The better idea is to do as much as computation as you can...

wouldn't it be just better to fix SparseVectorView + SparseVectorView addition? @ViralBShah yet another argument for `AbstractCompressedSparseVector`?

@ViralBShah so do i extract the AbstractCompressedVector from #201 and merge it? We can make the default for all operations on `AbstractCompressedVector`s return `SparseVector`s

```julia +(x::SymPy.SymbolicObject, y::SymPy.SymbolicObject) in SymPy at ~/.julia/packages/SymPy/5GXQf/src/mathops.jl:9 Body::Any 1 ─ nothing │ 9 2 ─ %2 = SymPy.getfield(x, :__pyobject__)::PyCall.PyObject │╻╷╷ getproperty │ %3 = invoke PyCall.__getproperty(%2::PyCall.PyObject, :__add__::Symbol)::PyCall.PyObject ││╻ getproperty │...

there seems to be an assumption in a few places (including `==`) that all zeros are equal. maybe this assumptions should be relaxed to only `Real` types? i.e. if the...