ForwardDiff.jl
ForwardDiff.jl copied to clipboard
JuliaDiff
Forward Mode Automatic Differentiation for Julia
#274 fixed an issue with `muladd` which only came up with tags were in play. The fix included adding a non-`Void` tag to every dual in the tests. In doing...
IIUC, ForwardDiff.jl does not support complex valued function such as `x->exp(im*x)`. Is there any plan to support this? I believe this is possible since [AutoGrad.jl](https://github.com/denizyuret/AutoGrad.jl) seems to support this.
Follow-up issue to #179. From my comment there: > In a future PR, we could add `AbstractConfig` constructors like `GradientConfig(x, Chunk{N}, NaNSafe{true})`. The generated dual numbers would then contain partials...
My apologies if this has been suggested before: While ForwardDiff is not the heaviest of packages in the ecosystem, it's also not exactly lightweight (take 1.6 seconds to load on...
Noone wants to change FowardDIff. It is stable. Even if one *does* want to change ForwardDiff, noone wants to change the API in a breaking way.
This implements the `Partials` arithmetic with `Complex` numbers in addition to `Real`, and enables the retrieval of `partials` from `Complex{ C^m works, as well as the in-place variation `jacobian!`. Tests...
Consider this function: ```julia sq(x) = x==1 ? one(x) : x^2 @test FiniteDifferences.central_fdm(5, 1)(sq, 1) ≈ 2.0 @test_broken ForwardDiff.derivative(sq, 1.0) == 2.0 ``` Here `ForwardDiff` gets the wrong answer, according...
When trying to differentiate the numerical integral from QuadGK (here integral of sin(x) for example), I get a stackoverflow error. Differentiating with Zygote works. ```julia using ForwardDiff using QuadGK f(x)...
…rectly the function Right now duals of duals of duals have an exponential growth in the type name because the functions then to have closures which hold values with the...
Hi! I am having trouble differentiating a function with ForwardDiff.jl. I don't know why, but the resulting jacobian contains `NaN`s. Below is a MWE to kickstart the discussion. Consider the...
