8000 Add "scoring" argument to ``score`` · Issue #28995 · scikit-learn/scikit-learn · GitHub
[go: up one dir, main page]

Skip to content
Add "scoring" argument to score #28995
Open
@amueller

Description

@amueller

Describe the workflow you want to enable

I want to enable non-accuracy metrics to estimator.score, and ultimately deprecate the default values of accuracy and r2. I would call it scoring though it's a bit redundant but consistent.
That would allow us to get rid of the default scoring methods, which are objectively bad and misleading, and it would require the minimum amount of code changes for anyone.

Describe your proposed solution

Replace

est.score(X, y)

with

est.score(X, y, scoring="accuracy")

or rather

est.score(X, y, scoring="recall_macro")

(or r2 for regression).

Describe alternatives you've considered, if relevant

  • Keep current status, which is bad (both accuracy and r2 are bad)
  • Remove scoring method.

I can't think of any other alternatives tbh.

Additional context

I think in theory this requires a slep, as it's changing shared API, right?

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions

      0