-
Hi,
I have a question about using the automatic differentiation to get the local sensitivity by an adaptive ODE solver. For some ODE solvers, the stepsize might depend on the parameters of the ODE sy…
-
```jl
using OrdinaryDiffEq
using DiffEqSensitivity
function fb(du,u,p,t)
du[1] = dx = p[1]*u[1] - p[2]*u[1]*u[2]*t
du[2] = dy = -p[3]*u[2] + t*p[4]*u[1]*u[2]
end
p = [1.5,1.0,3.0,1.0]; u0…
-
## Description
I think there's a bug in `set_zero_all_adjoints_nested` where we zero one `vari` from the stack above.
Test code is:
```cpp
TEST(AgradRev, chainable_object_nested_testadj) {
…
-
```julia
using SpecialFunctions
using Zygote
Zygote.gradient(besseli, 4, 0.3)
```
yields
```julia
ERROR: not implemented
Stacktrace:
[1] error(s::String)
@ Base ./error.jl:33
[2] …
-
## 🚀 Feature
As per title
## Motivation
Much needed for backward methods with complex support in forward.
cc @ezyang @anjali411 @dylanbespalko @vincentqb @vishwakftw @jianyuh @nikitaved @p…
-
In the formulation of the objective functional, the L1 regularization dampings do not appear. They also are not in the constrained problem. They appear only on the iteration algorithm. Is this correct…
-
Hello it is unfortunately not possible to use automatic differentation with (at least) the `MvNormal` distribution.
The following code will fail at `rand(p)` due to a wrong conversion to `Float64`
`…
-
In https://arxiv.org/abs/2005.13420 Onken and Ruthotto conclude using some of the examples in DiffEqFlux that Disc-Opt methods can achieve similar performance to Opt-Disc at inference with drastically…
-
just to file an issue that will prob resolve https://github.com/chriselrod/StrideArrays.jl/issues/21 so that I can just depend on this package.
-
Zygote currently fails to differentiate through `kerneldiagmatrix` when given a `RowVecs` or a `ColVecs`
MWE :
```julia
using KernelFunctions, Zygote
X = KernelFunctions.RowVecs(rand(3, 3))
…