Hi all, I’m currently using Convex.jl as one of the packages to solve an optimisation problem. I was wondering if anyone may able to offer any insight into the issue I’m facing.

Currently, I am trying to perform a Lasso regression on a Matrix of vectors for example:

|[a,b] [c,d]|

|[e,f] [g,h]|

I am currently following the example function of LassoEN found in the documentation of the package and it does work perfectly fine for a standard matrix of scalars.

|a b|

|c d|

Upon running on a matrix of vectors I obtain the following error:

Element type mismatch. Tried to create an `Adjoint{Adjoint}`

from an object with eltype `Array`

, but the element type of the adjoint of an object with eltype `Array`

must be `Adjoint{_A, _B} where {_A, _B}`

.

Stacktrace:

[1] error(s::String)

@ Base .\error.jl:33

[2] checkeltype_adjoint(#unused#::Type{Adjoint}, #unused#::Type{Array})

@ LinearAlgebra C:\buildbot\worker\package_win64\build\usr\share\julia\stdlib\v1.6\LinearAlgebra\src\adjtrans.jl:76

[3] Adjoint{Adjoint, Vector{Array}}(A::Vector{Array})

@ LinearAlgebra C:\buildbot\worker\package_win64\build\usr\share\julia\stdlib\v1.6\LinearAlgebra\src\adjtrans.jl:38

[4] Adjoint(A::Vector{Array})

@ LinearAlgebra C:\buildbot\worker\package_win64\build\usr\share\julia\stdlib\v1.6\LinearAlgebra\src\adjtrans.jl:95

[5] adjoint(A::Vector{Array})

@ LinearAlgebra C:\buildbot\worker\package_win64\build\usr\share\julia\stdlib\v1.6\LinearAlgebra\src\adjtrans.jl:136

[6] top-level scope

@ REPL[102]:1

If anyone has any insight into my issue any help would be greatly appreciated, Thank you in advance for your help! Also please let me know if my problem needs a better explanation, I tried my best to show what I meant by a matrix of vectors as opposed to a matrix of scalars. I can elaborate more if needed!