Interface for LowRankModels.jl

[tlienart]

(I know there's still #35 waiting to be done but the models for which there already is a ScikitLearn.jl interface are somewhat easier to write interfaces for which, hopefully, will make me confident enough to help out on more complex stuff later).

LowRankModels.jl implements quite a lot of interesting stuff:

• PCA (iterative approach so should be seen as "approxPCA" afaik)
• QPCA (quadratically regularised PCA)
• NNMF (non neg matrix facto)
• KMeans (note they explicitly say it's not the most efficient way of doing KMeans so while we may want to offer an interface for it, there may be something better later [1])
• RPCA (robust PCA)

[1] how do you envisage competing algorithms for the same task? E.g. if two packages do KMeans and implement MLJBase? Or, in fact, if two packages implement PCA like here and MultivariateStats?

I'm happy to try implement an interface for these.

[ablaom]

That would be wonderful!!

Regarding competing algorithms, let me continue the discussion here

[ablaom]

Just a warning. If you are implementing transformers rather than supervised models (e.g. PCA) then there is no coerce method and the user may present the input X to fit/predict in any Query.jl iterable table format. Ideally, transform will then be an object of the same data type (assuming it is a supported sink type). So you will need some familiarity with Query.jl. I am currently working on some more MLJBase utilities to make this less painful but WIP.

[tlienart]

Ok thanks, I will try and might again need some guidance in a WIP PR. I thought I'd mimic some of the stuff that you looked at for MultivariateStats

[ablaom]

I only implemented RidgeRegressor from there.

[ablaom]

The transformers we have live in src/builtins/Transformers.jl, the best example is Standardizer. However, it is not in a good state but I am working on that right now.