Valentin Churavy

You can also specify the inheritance tree with `abstract type`s and then have a `ParameterSet{Earth}`, where `Earth` is a `abstract type`. Not sure what the best solution here is.

``` spdiagm(kv::Pair{

``` julia> using SparseMatricesCSR [ Info: Precompiling SparseMatricesCSR [a0a7dd2c-ebf4-11e9-1f05-cf50bc540ca1] julia> using SparseArrays julia> SparseMatrixCSR(transpose(spdiagm(1=>ones(10)))) 11×11 SparseMatrixCSR{1, Float64, Int64} with 10 stored entries:Error showing value of type SparseMatrixCSR{1, Float64, Int64}: ERROR:...

So this looks like inference deciding that it can't resolve that call to a precise method instance.

Ah! I was looking at a talk proposal and without a tag being defined the "Tags (Optional):" Field is not visible and so I hadn't discovered that tags indeed already...

I think the canonical way is something like https://github.com/JuliaLang/julia/commit/f106bd9708fd61735f658e3e27ee1f227b829736

Can you show some examples where this is useful and how the format ended up? Would it be possible to add tests?

> That's still in scope, but again, not addressed by this. I think the cleanest way to address that is to also make CodeInstance parametric, so that the .inferred field...

I don't yet see the end design and how it will work. It may be that an example of a not `DefaultSpec` would be helpful for me to understand the...

The https://github.com/cesmix-mit/PotentialLearning.jl/blob/77fdd82c0ae3556c8d3f4165cba45194f2840629/Project.toml#L30 section is indeed quite bare. `Julia 1.9` now has `]compat` to easily edit the section, but we should encode the versions of dependencies we require.