v0.3.7

MLJGLMInterface v0.3.7

Diff since v0.3.6

• (enhancement) Models can now handle categorical inputs (predictors) (#45) @tiemvanderdeure

Merged pull requests:

• Retain data in tabular form (#45) (@tiemvanderdeure)
• Add generic interface tests from MLJTestInterface.jl (#46) (@ablaom)

v0.3.6

MLJGLMInterface v0.3.6

Diff since v0.3.5

• Expose raw GLM models in reports (#41)

Merged pull requests:

• Expose raw fitted GLM model in report (#41) (@ablaom)

Closed issues:

• RFC: Return fitted model as fitresult (#40)

v0.3.5

MLJGLMInterface v0.3.5

Diff since v0.3.4

Closed issues:

• Should we also make the p-values available? (#36)

Merged pull requests:

• Explicitly add StatsModels dependency (#37) (@rikhuijzer)
• Slight improvements to report_keys documentation (#39) (@rikhuijzer)

v0.3.4

MLJGLMInterface v0.3.4

Diff since v0.3.3

Closed issues:

• GLM compat broken (#34)

Merged pull requests:

• Fix wrong formula order (#35) (@rikhuijzer)

v0.3.3

MLJGLMInterface v0.3.3

Diff since v0.3.2

Closed issues:

• Doc string tweaks (#32)

Merged pull requests:

• Bump MLJModelInterface to 1.4 and remove descr declarations (#33) (@ablaom)

v0.3.2

MLJGLMInterface v0.3.2

Diff since v0.3.1

Closed issues:

• Documentation tracking issue! (#25)
• Update the doc-strings (#30)

Merged pull requests:

• Add docstrings! (#26) (@josephsdavid)
• Add doc-strings (#29) (@ablaom)
• Update docstrings to reflect changes in the master branch (#31) (@josephsdavid)

v0.3.1

MLJGLMInterface v0.3.1

Diff since v0.3.0

Closed issues:

• Inputs that are wrapped matrix adjoints not working (#27)

Merged pull requests:

• Allow AbstractMatrix in augment_X (#28) (@rikhuijzer)

v0.3.0

MLJGLMInterface v0.3.0

Diff since v0.2.0

Closed issues:

• StackOverflowError in GLM.GlmResp (#8)
• Fitresult of GLM models contain data! (#16)

Merged pull requests:

• Set minimum version to Julia 1.6 (#18) (@rikhuijzer)
• Remove Parameters.jl dependency and refactor (#19) (@OkonSamuel)
• fix bug in LinearRegressor model fit (#20) (@OkonSamuel)
• Introduce a new fit return object FitResult (#21) (@OkonSamuel)
• allow report customization with report_keys kwarg. (#22) (@OkonSamuel)
• Add support for observation weights (#23) (@OkonSamuel)
• add more parameters to control fitting, and add data checks (#24) (@OkonSamuel)

v0.2.0

MLJGLMInterface v0.2.0

Diff since v0.1.7

Closed issues:

• Allow fitting arbitrary @formulas (#13)

Merged pull requests:

• fix bug in requesting keys of table sources (#14) (@OkonSamuel)

v0.1.7

MLJGLMInterface v0.1.7

Diff since v0.1.6

• Include feature names in fitted parameters (#12) @rikhuijzer

Closed issues:

• Issue to trigger new releases (#1)
• Why is LinearRegressor Probabilistic? (#9)
• Constants in linear model fit (#11)

Merged pull requests:

• Report feature names in fitresult (#12) (@rikhuijzer)