-
Hi,
I'm trying to modify [the heterogeneous test case on the GPU](https://enzyme.mit.edu/getting_started/CUDAGuide/#heterogeneous-ad) to get the forward mode only.
I first try the example as it…
-
I presume you aren't aware of this,
https://github.com/rdyro/torch2jax
How does it compare to your work?
-
### Description
```python
import jax
import jax.numpy as jnp
@jax.custom_vjp
def f(x, y):
return jnp.sin(x) * y
def f_fwd(x, y):
return f(x, y), (jnp.cos(x), jnp.sin(x), y)
def f_…
-
This works on Julia 1.9 and Enzyme main (6692fad483dbe4002be9d254dae959c634d841c9):
```julia
using Enzyme
autodiff(Reverse, x -> sum(skipmissing([x, missing, 3.0])), Active, Active(1.0))
```
```
…
-
MWE:
```julia
using Enzyme, ForwardDiff
using LinearSolve, LinearAlgebra, Test
n = 4
A = rand(n, n);
dA = zeros(n, n);
dA2 = zeros(n, n);
b1 = rand(n);
db1 = zeros(n);
db12 = zeros(n);
…
-
The following Julia code attempts to differentiate a convolution from NNlib but fails:
```julia
using Enzyme
Enzyme.API.runtimeActivity!(true)
using NNlib
w = randn(Float32, 3, 3, 5, 7);
dw …
-
I get an error using Enzyme when there is a sparse Cholesky back substitution. My guess is that this is just a missing rule? A mwe is
```julia
using Enzyme
using SparseArrays, LinearAlgebra
# …
-
## 🐛 Bug
Hi all I'm trying to build Caffe2 on Pytorch from Binaries, have used Python 2.7 with VS 2019 and CUDA 10.1. Cmake donwloaded manually and also updated via Conda as read from other users p…
-
@wsmoses Following #1152 's fix, I now get the following error when I try to differentiate the following code:
```julia
using Redbird
using SparseArrays
using MATLAB
using Enzyme
###############…
-
I'm playing with `Enzyme.jl` on an [IPU](https://www.graphcore.ai/products/ipu). Autodiff of a few functions work out-of-the-box, which is awesome, but differentiation of `log` chokes:
```julia
f(x…