-
I know that the adjoint source is defined as the derivative of objective function with respect to the electric field. If the objective function is defined as the square of a (a is mode coefficient), t…
-
Hi there,
How long is the umxACE run time usually? I am running it on my laptop with 4 cores with 8GB RAM on microbiome data (OTU counts) after umx_residualize and umx_scale_wide_twin_data. It has …
-
If F ⊣ G and F ⊣ H, then G ≅ H, and similarly on the other side.
-
It could be that I didn't read the documentation carefully, but does DifferentialEquations.jl support solution of the second order sensitivity equations? Second order derivatives with respect to param…
-
I was surprised to find today that the `checkpoint` argument ([docs here](https://docs.sciml.ai/stable/analysis/sensitivity/#Sensitivity-Algorithms-1)) to the `BacksolveAdjoint` sensealg has no real e…
-
There's a lot of parts of the documentation that can be better described:
- [ ] MNIST example can get a lot more text on what it's doing. Especially how it's performing minibatching.
- [ ] Every t…
-
```julia
using Zygote
f(args...;sensealg=nothing,kwargs...) = g(sensealg,args...;kwargs...)
g(args...;x=1,save_idxs=Colon(),kwargs...) = x[save_idxs]
Zygote.gradient(x->sum(f(;x=x,save_idxs=1:1)),…
-
I'm working on automatic differentiation at the level of compute expressions, and I would like to share some progress and hear any comments. Currently the automatic differentiation works well enough …
-
Sorry for fuzzy issue, but I'm a bit uncertain what is desired behaviour in this case.
Issue I'm seeing is this:
```julia
julia> loss(x,y) = Flux.mse(Chain(MaxPool((1,1)), z -> Flux.selu.(z))(x…
-
I want to do an adjoint definition like:
```julia
ZygoteRules.@adjoint function (f::ODEFunction)(u,p,t)
if f.vjp === nothing
ZygoteRules.adjoint(f.f,u,p,t)
else
f.vjp(u,p,t)
end…