JuliaAI / JuliaAI/ScientificTypes.jl
Is it possible to specify the number of classes when coercing into `Multiclass`?
Nobody has claimed this yet.
- Dominant language
- Julia
- Stars
- 98
- Forks
- 7
- PR merge metrics
- No merged PRs in 30d
Description
As per title: let's say I have a dataset d1 with a variable X which is Multiclass{10}, but now I get another smaller dataset d2 in which X only has 7 of the 10 classes present in d1. If I just coerce(d2, :X => Multiclass) in then it will be a Multiclass{7} and then downstream transformations/functionality in things like the MLJ.Pipeline will break. Can I specify the number of classes in Multiclass when coercing, along with how to assign a class, such that X in d2 could be coerced to the same scitype as X in d1?
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reviewing the coerce API for Multiclass and how its class count is determined. Use the d1/d2 scenario in the issue to define how a target class count and class assignment should work, then verify that the resulting scitype remains compatible with downstream MLJ.Pipeline functionality.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- julia
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100