HI, i’m looking to implement an implace hessian of a vector-valued function, for example: f(x) = [x[1],x[2],sin(x[1]),sum(x)] The ForwardDiff Documentation gives an out of place example (slightly wrong, as size(f(x)) !=size(x) in the general case: julia> function vector_hessian(f, x) n = …

Although I didn’t use the same method, the idea is here, to pass a implace Jacobian to another implace Jacobian, the cache needs to be build dinamically, and the inner jacobian should be passed as an out of place cached function.

And for reference, FiniteDiff.jl has a fully cached version of Hessian computations through finite differencing: [image] GitHub - JuliaDiff/FiniteDiff.jl: Fast non-allocating calculations of... Fast non-allocating calculations of gradients, Jacobians, and Hessians with sparsity support …

Nice! I also need to implement this on FiniteDiff (as an user selected option)

But the Hessian computations in FiniteDiff.jl do no support vector-valued functions, right?

Implementation of an implace Hessian of a vector-valued function?

General Usage

tkoolen January 24, 2020, 3:13am 2

See Nested ForwardDiff.jacobian calls with inplace function - #3 by tkoolen for one solution.

Topic		Replies	Views
Hessian of a vector-valued in-place function with ForwardDiff.jl Modelling & Simulations question	2	73	January 6, 2026
Nested ForwardDiff.jacobian calls with inplace function General Usage	6	1647	March 3, 2019
ForwardDiff.derivative! of inplace functions Performance forwarddiff	1	639	July 5, 2021
Hessian-vector product in place with ReverseDiff General Usage differentiation	7	1156	April 9, 2020
Repeatedly calculating a hessian using both ForwardDiff and ReverseDiff Optimization (Mathematical) question , differentiation	3	1568	February 2, 2021

Implementation of an implace Hessian of a vector-valued function?

Related topics