St. Elmo Wilken
St. Elmo Wilken
Chiming in here, this patch solved [this issue for me that involved sparse arrays](https://discourse.julialang.org/t/incorrect-jacobian-using-forwarddiff-and-sparse-functions/79901). I would offer to help, but unfortunately I just don't know enough about auto diff :/
Hi there, will this make it into master before 1.0?
sure thing, here it is: ``` julia> jac = Enzyme.jacobian(Forward, F, z) # crash o: %7 = call cc37 nonnull {} addrspace(10)* bitcast ({} addrspace(10)* ({} addrspace(10)*, {} addrspace(10)**, i32)*...
This worked! Thanks :) Although I ran into another issue, but I am not sure if they are related: ```julia using Enzyme using SparseArrays, LinearAlgebra using Random N = 10...
Sure thing, I can supply the model too if you like. To check if the SBML is valid I use [the sbml.org group's validation tool](http://sbml.org/Facilities/Validator/index.jsp). However, the model does load...
Some recently released cobra software, e.g. [COBREXA](https://github.com/LCSB-BioCore/COBREXA.jl), make use of the latest SBML standards, and this is actually what led me here :)
This works 🥳 @ChrisRackauckas and @shashi, thank you both for helping me so quickly!
Sorry to be a squeaky wheel, but I just noticed that although the patch works (🎊) it is substantially slower than the code that just lumps everything into a vector,...
I also ran into this issue! I have ```julia f(x)' * y ``` where `f` is a julia function yielding a sparse matrix.
Thanks to both of you for fixing these issues. I really appreciate the effort you put into this package, it is great!!