Daniel O'Malley

This is causing problems for me as well. Zygote gradients of other things like `f_exp(x) = sum(real(exp.(x)))` also return `nothing` for complex `CuArray`s, similar to issue #961. That makes me...

BTW, I added a [GitHub action to run the tests](https://github.com/OrchardLANL/DPFEHM.jl/actions/runs/2632551544).

Regarding Wu Hao and Dylan Harp, I got the author list by including the people who made a commit. If I do `git log | grep Ha`, I see Dylan...

@WilkAndy No, your suggestions are good. 😄 I've just been busy. I'm planning to get back to this next week.

@WilkAndy, I addressed the issues raised [here](https://github.com/openjournals/joss-reviews/issues/4544#issuecomment-1178305351), following the outline [here](https://github.com/openjournals/joss-reviews/issues/4544#issuecomment-1178365752). I also looked into all the issues raised in the repo and got the code coverage up over 85%...

Here's an updated version with `using GenericSVD` that produces a StackOverflowError instead of the MethodError in the old stack trace: ``` using Test import AlgebraicMultigrid import ForwardDiff using GenericSVD import...

I'm working on it.

Awesome work getting this rolling, @ranjanan and @DhairyaLGandhi! I was able to get Ranjan's example working on my machine. However, I run into an issue for larger problems (the example...

I'm basically in the same boat as @darsnack. I have a complex code that doesn't define `∇pmap`. I thought that discourse thread was a quick and easy way to provide...