[Tracking] Updating dependencies to CategoricalArrays 1.0

[ablaom]

CategoricalArrays 1.0 breaks a few things in MLJ. See here for some details. Here's the rough sequence for rolling these out:

• [x] Update CategoricalDistributions.jl and tag as breaking, v0.2.0
• [x] Update ScientificTypes.jl
• [x] Update StatisticalMeasures.jl
• [x] Update StatisticalMeasuresBase.jl
• [x] Update MLJBase.jl https://github.com/JuliaAI/MLJBase.jl/pull/1028
• [x] MLJ https://github.com/JuliaAI/MLJ.jl/pull/1185

I will do these.

Then, in no particular order, we should review the compat requirements for the following core packages. I do not expect breakages but in a few cases we might start getting dep warnings which would be good to address these early.

• [x] MLJTuning.jl Does not appear to need updating (tested with a draft MLJBase update)
• [x] MLJIteration.jl ditto
• [x] MLJEnsembles.jl https://github.com/JuliaAI/MLJEnsembles.jl/pull/42
• [x] MLJFlux.jl (needed for MLJTransforms doc generation) https://github.com/FluxML/MLJFlux.jl/pull/311 passes tests, docs failing
• [x] MLJTransforms (needed for testing in MLJModels) https://github.com/JuliaAI/MLJTransforms.jl/pull/61 local tests: ✔️
• [x] MLJMultivariateStatistics (needed for MLJModels testing) https://github.com/JuliaAI/MLJMultivariateStatsInterface.jl/pull/63 local tests: ✔️
• [x] MLJDecisionTreeInterface.jl (needed for tests in FeatureSelection.jl and MLJModels.jl) https://github.com/JuliaAI/MLJDecisionTreeInterface.jl/pull/60 local tests: ✔️
• [x] MLJModels.jl (needed for tests in MLJBalancing) https://github.com/JuliaAI/MLJModels.jl/pull/597 local tests: ✔️
• [x] Imbalance.jl (needed for testing MLJBalancing) https://github.com/JuliaAI/Imbalance.jl/pull/105
• [x] MLJBalancing.jl (doesn't appear to need code changes, but depends on MLJBase)
• [x] FeatureSelection.jl https://github.com/JuliaAI/FeatureSelection.jl/pull/30 local tests: ✔️
• [x] ARFFFiles.jl (needed by OpenML, on which MLJ depends) https://github.com/cjdoris/ARFFFiles.jl/pull/26 local tests: ✔️
• [x] Maxnet.jl https://github.com/tiemvanderdeure/Maxnet.jl/issues/28

Here are some other packages I've checked and don't appear to need attention:

• MLJScikitLearnInterface

Additional action required after MLJ itself is updated:

• [x] https://github.com/JuliaAI/MLJTransforms.jl/issues/62 (MLJTransforms)
• [ ] https://github.com/FluxML/MLJFlux.jl/issues/312 (MLJFlux)

cc @DilumAluthge @devmotion

[ablaom]

And then, some packages providing implementations of the MLJ API will likely need updating to CategoricalArrays 1.0 and/or CategoricalDistributions 0.2:

• [x] BetaML
• [x] CatBoost (requested)
• [x] EvoTrees
• [ ] LaplaceRedux (requested)
• [x] MLJLIBSVMInterface
• [ ] OneRule (requested)([issue posted]
• [x] OutlierDetectionInterface (requested)
• [x] SIRUS (requested)

The steps for updating are:

1. update the compats
2. change in /src and /test any reference to CategoricalArrays.levels(x) to unwrap.(CategoricalArrays.levels(x)) if necessary (or if you are not sure). Look out for postfix applications like x |> levels. Possibly one has to do something similar for any references to CategoricalArrays.unique but I haven't encountered a problem with that one yet.
3. change any reference to MLJBase.classes or MLJModelInterface.classes to CategoricaArrays.levels (or CategoricalDistributions.levels)
4. check tests

[juliohm]

Bumped into this issue today in one of our test suites. MLJ.jl is holding CategoricalArrays.jl < v1

[ablaom]

@juliohm MLJ is now updated. Use MLJ 0.22. It's only tagged breaking because of the change in behaviour of levels/unique; see the release notes for details.