-
```julia
using Enzyme, SparseArrays, Statistics
function bc0_test_function(ps)
z = view(ps, 26:30)
C = Matrix{Float64}(undef, 5, 1)
C .= z
return C[1]
end
@noinline functio…
-
So far we have only worked on forward mode autodiff. However, to implement backprop and build some simple machine learning application we'll need to implement some reverse mode autodiff.
For now I'…
-
As discussed on Slack, it is sometimes annoying to use reverse mode because you need `autodiff_thunk` whenever
- the function output is a number and the cotangent is different from `1.0`
- the funct…
-
I tried running the last example in the README with an HMC sampler substituted for the MALA one:
```
using Klara
plogtarget(z) = -dot(z, z)
p = BasicContMuvParameter(:p, logtarget=plogtarget…
-
In the notebooks on the documentation here:
https://jax.readthedocs.io/en/latest/notebooks/autodiff_cookbook.html
The author describes topics that he would like to showcase in a future Autodiff co…
-
An implementation of forward-over-reverse, i.e. applying forward mode autodiff to the result of reverse autodiff, can be a way to exploit AD for the computation of Hessians. It might make sense to dir…
-
I'm writing a custom rule for a function that takes, among other args, an Active tuple/namedtuple. This throws an error when the values in the tuple are taken from arrays at indices determined at runt…
-
```julia
using Enzyme
function f_ip(x, tmp)
tmp .= x ./ 2
return dot(tmp, x)
end
function f_gradient_deferred!(dx, x, tmp)
dtmp = make_zero(tmp)
autodiff_deferred(Reverse, …
-
The Enzyme extension only implements reverse rules for integrands whose activity is `Union{Const,Active}`. In my problem, the integrand wraps an ApproxFun.jl Chebyshev interpolant specified by a `Cons…
-
**Describe the bug**
The following Swift code causes a compiler crash.
**To Reproduce**
Steps to reproduce the behavior:
1. Copy the following reproducer into a Swift file on your computer, name…