v0.2.0

MLJDecisionTreeInterface v0.2.0

Diff since v0.1.4

• (mildly breaking) Make feature names encountered in training available in fitted_params and report (#14)
• (breaking) Remove pdf_smoothing option from all classifiers (#16)

Closed issues:

• Include feature names in the fitresult for use by external tree visualisation tools (#13)

Merged pull requests:

• Expose feature names (#14) (@ablaom)
• Remove pdf_smoothing option (#16) (@ablaom)
• For a 0.2.0 release (#17) (@ablaom)

v0.1.4

MLJDecisionTreeInterface v0.1.4

Diff since v0.1.3

Closed issues:

• Issue to trigger new releases (#1)
• why do input scitypes not include Multiclass? (#10)

Merged pull requests:

• Update doc-strings to meet new standard (#12) (@ablaom)
• Bump 0.1.4 (#15) (@ablaom)

v0.1.3

MLJDecisionTreeInterface v0.1.3

Diff since v0.1.2

• (enhancement) Expose the rng hyperparameter of DecisionTreeClassifier, DecisionTreeRegressor, RandomForestClassifier, and RandomForestRegressor (#4)

Closed issues:

• passing rng to fit (#4)

Merged pull requests:

• Expose rng addressing #4 (#9) (@ablaom)

v0.1.2

MLJDecisionTreeInterface v0.1.2

Diff since v0.1.1

• extend compat for MLJModelInterface

Merged pull requests:

• CompatHelper: bump compat for "MLJModelInterface" to "1.0" (#8) (@github-actions[bot])

v0.1.1

MLJDecisionTreeInterface v0.1.1

Diff since v0.1.0

Merged pull requests:

• Move from Travis CI to GitHub Actions CI (#3) (@DilumAluthge)
• Fix some invalid tests (#5) (@ablaom)
• update compat helper (#6) (@ablaom)

v0.1.0

MLJDecisionTreeInterface v0.1.0