-
This issue is to provide a minimal example of neural network training with Lux to hopefully make it easier for developers to work toward making it viable. It probably isn't news to anyone here that t…
-
Hi @gdalle,
Awesome package, thanks for making it!
I am considering whether to use this as the backend interface of [SymbolicRegression.jl](https://github.com/MilesCranmer/SymbolicRegression.jl)…
-
Inplace BatchNorm seems to be developed by Mapillary here: https://github.com/mapillary/inplace_abn
This would be a very nice addition to core PyTorch (for memory savings).
cc @ezyang @gchanan @…
-
Hi,
I am trying to integrate a vector field with jumps in the derivative and experience a high integration error when the derivative jumps. As a solution attempt, I would like to split the integrat…
-
In the spirit of creating lightweight interface-defining packages (see TuringLang/Bijectors.jl#199 which resulted in [InverseFunctions.jl](https://github.com/JuliaMath/InverseFunctions.jl) and [Change…
-
Was trying out the newly added Hessian-vector product function `hvp!` on a simple quadratic and I have run into the following strange error:
```julia
julia> using Enzyme: hvp!
julia> n = 10
…
-
I'm running Enzyme in reverse mode on some code on Julia 1.10.2. This however results in a segmentation fault with the following stack trace:
[out.txt](https://github.com/EnzymeAD/Enzyme.jl/files/145…
-
This might be a documentation request, or a feature request.
Currently the AutoMALA constructor takes the argument `default_autodiff_backend`, which it uses via LogDensityProblemsAD to differentiat…
-
By "vector mode" I mean the ability to propagate a chunk of seeds in the pullback, similar to ForwardDiff's `Chunk` or Enzyme's `BatchDuplicated`.
Thanks to https://github.com/gdalle/DifferentiationI…
-
As seen on CUDA.jl#master:
```
extensions/enzyme (2) | failed at 2024-08-22T19:02:35.224
Testing finished in 4 minutes, 37 seconds, 205 milliseconds
Worker 2 failed running test extensi…