v0.3.4

StatisticalMeasures v0.3.4

Diff since v0.3.3

• (new measure) Add the Continuous Boyce Index, as proposed by Hirzel et al (2006) (#51, #52)

Merged pull requests:

• add continuous boyce index (#52) (@tiemvanderdeure)
• Bump actions/checkout from 4 to 6 (#65) (@dependabot[bot])
• For a 0.3.4 release (#66) (@ablaom)

Closed issues:

• continuous boyce index (#51)

v0.3.3

StatisticalMeasures v0.3.3

Diff since v0.3.2

Merged pull requests:

• Fix reference to old code (#59) (@ablaom)
• Fix and properly test case of empty yhat in check_l2 (#60) (@ablaom)
• For a 0.3.3 release (#61) (@ablaom)

Closed issues:

• cross_entropy leads to ERROR: UndefVarError: isinvalid not defined in StatisticalMeasuresBase (#58)

v0.3.2

StatisticalMeasures v0.3.2

Diff since v0.3.1

Merged pull requests:

• Extend compat LearnAPI = "1, 2" (#56) (@ablaom)
• For a 0.3.2 release (#57) (@ablaom)

v0.3.1

StatisticalMeasures v0.3.1

Diff since v0.3.0

• Add Willmott index of agreement, willmott_d = WillmottD() (#53)

Merged pull requests:

• Add Willmott index of agreement (#53) (@marcom)
• For a 0.3.1 release (#54) (@ablaom)

v0.3.0

StatisticalMeasures v0.3.0

Diff since v0.2.1

• Bump the CategoricalArrays requirement to 1.0
• (mildly breaking) Arrange that StatisticalMeasures' overloading of the CategoricalArrays.levels function to confusion matrices mimics the new behavior: it returns a CategoricalVector, not an unwrapped version of the same.

Merged pull requests:

• Fixing typo in RSquared documentation (#41) (@pebeto)
• Update GitHub actions and add dependabot.yml (#42) (@abhro)
• Resolve corner case issue around misleading error message (#45) (@ablaom)
• Support CategoricalArrays 1.0 (#49) (@ablaom)
• For a 0.3 release (#50) (@ablaom)

v0.2.1

StatisticalMeasures v0.2.1

Diff since v0.2.0

Merged pull requests:

• Extend some compat bounds and dump pkg extension legacy code (#39) (@ablaom)
• For a 0.2.1 release (#40) (@ablaom)

v0.2.0

StatisticalMeasures v0.2.0

Diff since v0.1.7

• (breaking) Bump the LearnAPI compatibility bound to "0.2", which requires all references to LearnAPI.LiteralTarget to be replaced with LearnAPI.Point (#31)
• Bump the julia compatibility bound to "1.10", the current Long Term Support version, which is necessary, because this is the lower bound for LearnAPI@0.2 (#31)

Merged pull requests:

• Update README.md (#30) (@ablaom)
• LearnAPI compat bump (#31) (@ablaom)
• For a 0.2 release (#33) (@ablaom)

v0.1.7

StatisticalMeasures v0.1.7

Diff since v0.1.6

• balanced_accuracy can now be called on a confusion matrix (#28)

Merged pull requests:

• Define balanced accuracy for confusion matrices (#28) (@tiemvanderdeure)
• For a 0.1.7 release (#29) (@ablaom)

v0.1.6

StatisticalMeasures v0.1.6

Diff since v0.1.5

Merged pull requests:

• For a 0.1.6 release (#24) (@ablaom)

v0.1.5

StatisticalMeasures v0.1.5

Diff since v0.1.4

Merged pull requests:

• performance optimizations for Confusion Matrix (#21) (@tiemvanderdeure)
• Fix can_report_unaggregated trait for LogScore and BrierScore (#22) (@ablaom)
• For a 0.1.5 release (#23) (@ablaom)

Closed issues:

• Add probability of correct classification (PCC) as an alias for balanced_accuracy (#15)