Anubhab Haldar

Would that works as so? `mpirun -np 6 py-spy record --output mpitestoptimizer.scope --format speedscope python optimize.py --subprocesses -r 25` It appears to be doing what I'd like it to, but...

`Error: No python processes found in process 2722 or any of its subprocesses` I think a key problem here is that all MPI processes are in a sense equivalent, and...

@benfred I can close this issue if there's no clear way to get this, or I can leave it open if you think support for this may be added eventually.

https://web.stanford.edu/group/fan/S4/lua_api.html If you look at the AddMaterial documentation here, I believe the off-diagonal elements are ignored and assumed to be zero.

Is there any timeline on a fix for this?

I just ran `@btime` on this with x::Vector{Float64} and I got 6 allocations. ```julia julia> x = rand(4000); julia> @btime AAPE($x) 10.800 μs (6 allocations: 93.89 KiB) 0.4158251611947962``` Are datasets...

I tried the example given and I get an error which I've seen elsewhere, but don't quite know how to interpret: ```julia julia> x = Dataset(rand(100, 3)); julia> AAPE.(x) ERROR:...

```julia julia> x = Dataset(rand(1000, 15)); julia> @btime AAPE.($vec(x)); 7.900 μs (3 allocations: 7.97 KiB) ``` Still not seeing it, maybe I'm missing something? This is on Julia 1.8.3.

My bad, I was referring to the original issue of 10 allocations per element of `Dataset`, which I'm not seeing here. The above code snippet does in fact give 0...

Roughly in order of priority: 1. What is the correct way of selecting the correct calls depending on the value of theta and eta? I tried macros and inline functions,...