New version: MLJBase v0.8.1

[JuliaRegistrator]

• Registering package: MLJBase
• Repository: https://github.com/alan-turing-institute/MLJBase.jl
• Created by: @ablaom
• Version: v0.8.1
• Commit: af24366ac483f7969600bb74e3dfc03b46e6990a
• Git reference: master
• Release notes:

• (Enhancement) Give built-in Ames data sets types with improved scitype interpretations (PR Register DrWatson: 0.1.0 #92)

• (Enhancement) Add method is_same_except_method(m1, m2, exceptions....) that asks if two MLJType objects m1, m2 have the same field values, with the exception of the field names (symbols) nominated in exceptions.

• (Enhancement) Define Static <: Unsupervised model subtype and default fit for transformers without learned parameters (PR Register ApproxFunSingularities: 0.0.1 #104, MLJ #291)

• (Enhancement) Add confusion_matrix and the following measures: auc, accuracy,
  balanced_accuracy, matthews_correlation, truepositive, truenegative, falsepositive, falsenegative, truepositive_rate, truenegative_rate, falsepositive_rate, falsenegative_rate, falsediscovery_rate, negativepredictive_value, precision, plus varied aliases. Do ?measures() for full list (PR Register New Package Agents: 1.1.0 #88)

• (Enhancement) Add aggregate measure trait and associated aggregate(v, measures) method. Anticipating planned change to MLJ (PR Register New Package Agents: 1.1.0 #88)

• (Enhancement) Add measures() and measures(conditions...) for querying the full list of available measures (PR Register New Package Agents: 1.1.0 #88)

• (Enhancement) Add roc_curve to compute the receiver operator characteristic in binary classification (PR Register New Package Agents: 1.1.0 #88)

• (Breaking) Add @load stub (PR Register ImageMorphology: 0.2.0 #101). Breaks MLJModels which previously owned this method.

• Add supports_online model trait and update_data fallback, in preparation for MLJ online learning support (PR Register QHull: 0.1.0 #102, MLJ #60)

[github-actions]

Your new version pull request met all of the guidelines for auto-merging and is scheduled to be merged in the next round.

---

If you want to prevent this pull request from being auto-merged, simply leave a comment. If you want to post a comment without blocking auto-merging, you must include the text [noblock] in your comment.

[julia-tagbot]

I've created release v0.8.1, here it is.