Skip to content

Releases: mlr-org/mlr3measures

mlr3measures 1.0.0

12 Sep 13:59
Compare
Choose a tag to compare
  • Added new measure linex (Linear-Exponential Loss).
  • Added new measure pinball (Average Pinball Loss).
  • Added new measure mauc_mu (Mu AUC).

mlr3measures 0.6.0

22 Jul 08:31
Compare
Choose a tag to compare
  • Added binary classification measures gmean and gpr.
  • Added multiclass mcc.

mlr3measures 0.5.0

05 Aug 12:45
6d2e6e9
Compare
Choose a tag to compare
  • Added some observation-wise loss functions: ae(), ape(), se(),
    sle(), and zero_one(),

mlr3measures 0.4.0

29 Sep 15:45
8802af3
Compare
Choose a tag to compare
  • New measures to calculate set similarities: jaccard and phi.
  • Many measures now support sample weights.
    Supported measures have $sample_weights set to TRUE in the measures
    environment.

mlr3measures 0.3.0

05 Oct 14:27
Compare
Choose a tag to compare
  • New measure prauc (area under precision-recall curve) (#22) implemented in
    package PRROC.
  • Removed dependency on orphaned package bibtex.

mlr3measures 0.1.3

01 Apr 20:56
Compare
Choose a tag to compare
  • Added a packageStartupMessage() advising against attaching this package.
  • Added new measure bbrier (binary Brier Score).
  • Added new measure mbrier (multi-class Brier Score).
  • Improved numerical stability for measure mcc.

mlr3measures 0.1.1

13 Dec 09:03
Compare
Choose a tag to compare
  • Added new measure bacc (Balanced Accuracy).
  • Fixed some tests which stochastically failed.
  • The name of the measure is now also stored in the meta data.

mlr3measures 0.1.0

29 Oct 08:55
Compare
Choose a tag to compare
v0.1.0

desc