NEWS.md

In MlBayesOpt: Hyper Parameter Tuning for Machine Learning, Using Bayesian Optimization

MlBayesOpt 0.3.3

• Fix Buggs for xgb_cv_opt()

MlBayesOpt 0.3.2

• Add a vignette
• Changed a option num_tree_range in rf_opt() to num_tree
• Fix Bugs for SVM functions

MlBayesOpt 0.3.1

Fixed bugs

MlBayesOpt 0.3.0

• Add a function for cross validation (SVM), svm_cv_opt().
• Add choices for SVM functions(svm_opt() and svm_cv_opt()). We can use linear, polynomial, radial(default), sigmoid kernels in svm_kernel option.

• We don't have to write the object name for specifying label column.

  • before:

  R res <- svm_opt(train_data = iris_train, train_label = iris_train$Species, test_data = iris_test, test_label = iris_test$Species, acq = "ucb" )

  • after:

  R res <- svm_opt(train_data = iris_train, train_label = Species, test_data = iris_test, test_label = Species, acq = "ucb" )

MlBayesOpt 0.2.1

Fix minor bugs

MlBayesOpt 0.2.0

Support for cross validation, xgb_cv_opt()

MlBayesOpt 0.1.2

Fixed minor bugs

MlBayesOpt 0.1.1

xgb_opt() supports objective function "binary:logistic"

MlBayesOpt 0.1.0

Delete all warnings and notes

MlBayesOpt 0.0.9300

Add rf_opt()

MlBayesOpt 0.0.9200

Add parameter range to the arguments

MlBayesOpt 0.0.9100

Add svm_opt()

MlBayesOpt 0.0.9000

The first release. xgb_opt() only.

MlBayesOpt documentation built on May 2, 2019, 9:20 a.m.