-
As noticed in https://github.com/google/jax/pull/3398:
```
In [9]: from jax import vjp
In [10]: out, f_vjp = vjp(lambda x: 1j * x, 1.0)
In [11]: f_vjp(1 + 0j) # wrong!
Out[11]: (DeviceArray(…
-
I'm interested in implementing autodiff based on expression templates. This would enable compile-time symbolic differentiation that reduces runtime cost. However, this may restrict runtime functionali…
-
From @compnerd’s comment in https://github.com/tensorflow/swift-apis/pull/1184
> The project doesn't seem particularly enticing to me - the "rebuilding from the ground up for GPU acceleration thro…
-
## Description
There are two ways to improve the way in which we compute the Jacobian matrix of an algebraic solver:
* compute the Jacobian of the system with respect to the parameters and the unkno…
-
When I run the following I get an error. I am on Julia 1.8.0, Enzyme 0.10.4 and CUDA 3.12.0.
```julia
using Enzyme, CUDA, StaticArrays
struct Atom{T}
σ::T
ϵ::T
end
function force(c1…
-
| | |
|------------------|-----------------|
|Previous ID | SR-15849 |
|Radar | None |
|Original Reporter | @BradLarson …
-
Expressions containing `pow(x,2)` can produce NaNs when computing a Hessian in reverse mode.
In the example of computing the gradient and Hessian for the Rosenbrock function below, the computed gradi…
-
Hello,
How does auto differentiation for a loss function work which itself is an outcome of an iterative process?
In particular, loss(x, a) = min_z f(z, x, a), where the min_z f can be solved wit…
-
I'm looking for a reverse mode autodiff package that supports in-place mutation and @ChrisRackauckas suggested this. Unfortunately it crashed on first attempt:
```julia
julia> using FillArrays, Band…
-
Is this package for deterministic or stochastic difference equations?
It would be great to fill both left quadrants
![image](https://user-images.githubusercontent.com/7883904/136105161-d1430f4e-0…