Anthony Blaom, PhD
Anthony Blaom, PhD
After - [ ] https://github.com/JuliaAI/MLJBase.jl/pull/1026 is merged Fix [this link](https://github.com/JuliaAI/MLJModelInterface.jl/blob/86a9845b061fb7bbb8a6b613c5bbb9c71ccadc6f/docs/src/fitting_distributions.md?plain=1#L18) to point to end of `resampling.jl` tests in MLJBase.
By "multitarget" I mean vector-valued target `y` for regression. In particular it would be good to have a multi-target version of elastic net regression, as I don't believe there is...
https://github.com/ablaom/MLJTutorial.jl/issues/39
**Edit**. It seems test pass on 1.10 but fail on 1.11. Multiple failures are documented below. As noted, it seems like the issue is the use of unstable RNGs in...
Some context: https://github.com/JuliaAI/MLJ.jl/issues/1172#issuecomment-3499831411
CategoricalArrays 1.0 breaks a few things in MLJ. See [here](https://github.com/JuliaAI/CategoricalDistributions.jl/pull/75) for some details. Here's the rough sequence for rolling these out: - [x] Update CategoricalDistributions.jl and [tag as breaking, v0.2.0](https://github.com/JuliaRegistries/General/pull/136725)...
We break up the very long `src/resampling.jl` file into separate files: ``` resampling ├── error_messages.jl ├── evaluate.jl ├── evaluation_results.jl ├── logging.jl ├── resampler.jl ├── strategies.jl └── train_test_pairs.jl ``` This PR...
But keep the method to ensure backwards compatibility with MLJModelInterface 1.0. In MLJBase 1.10 the method calls `CategoricalArrays.levels`. Previously it called `CategoricalDistributions.classes`.
## Context In an MLJ pipeline, components are trained in sequence. If a component is mutated (replaced or its hyperparameters changed) retraining the pipeline only triggers retraining of that component...
Currently MLJBase and ScientificTypes both export `scitype`, which creates a name clash. This has not appeared to have caused many problems because MLJBase, which has ScientificTypes as a dependency, re-exports...