Sam

Proposed transformation: Loading matrix for factor analysis

2

Would there be any interest in a "special array" transformation to create loading matrices for factor analysis? i.e., when `n` observed variables are represented by `m

NamedTuple parameters don't work with AutoModelingToolkit()

3

I defined an `OptimizationFunction` and `OptimizationProblem` which expect the parameters as a named tuple (which is convenient when there are a bunch of parameters, since they can be referred to...

Moving this question here from [Discourse](https://discourse.julialang.org/t/optimization-jl-interface-for-turing-models/107473). The `optim_problem` interface to Optimization.jl is currently not documented anywhere. Is this intentional (e.g. because the interface is a moving target), or would a...

A question/suggestion on this example: https://github.com/TuringLang/Turing.jl/edit/master/docs/_tutorials/10_bayesian-differential-equations.md It currently has the data normally distributed around the ODE solution. However, given that this is an ecological model, one would never actually observe...

Fix for method ambiguity in https://github.com/JuliaDiff/ForwardDiff.jl/issues/686. Couldn't tell if/how tests should be updated to reflect this change, happy to do that too if someone more knowledgeable can give me a...

promote_rule ambiguity with AbstractIrrational and ForwardDiff.Dual

1

Just ran into this error: ```julia ERROR: MethodError: promote_rule(::Type{IrrationalConstants.Twoπ}, ::Type{ForwardDiff.Dual{ForwardDiff.Tag{OptimizationReverseDiffExt.OptimizationReverseDiffTag, Float64}, Float64, 12}}) is ambiguous. Candidates: promote_rule(::Type{S}, ::Type{T}) where {S

Add more methods for basic summary stats for Chains

3

I was recently surprised to find that calling `std()` on a `Chains` threw an error. Looking into it, it appears that only a few individual summary statistics (`mean`, `cor`, and...

Dates/DateTime x-values converted to categorical in groupedbar

1

`groupedbar` produces a messy plot when given `Date`s as x-values. The bar widths can be off, and it plots a label for every unique x-value: ```julia using DataFrames, StatPlots x...

Just chased a type-instability in some code back to the `getproperty` method for `LDLFactorization`s [here](https://github.com/JuliaSmoothOptimizers/LDLFactorizations.jl/blob/cf1532ffb0a74b916208e085f42f925ae10bbc15/src/LDLFactorizations.jl#L849). Using `getfield` instead is a workaround, but it would be nice if the dot-property syntax...