ehymet 0.1.1
Major Changes
- Implemented automatic variable selection functionality
- Added new
vars_combinations = "auto" option in
EHyClus
- Integrated
findCorrelation and
select_var_ind functions in utils.R
- Enhanced clustering validation metrics
- Added Adjusted Rand Index (ARI) to
clustering_validation
- Added internal validations metrics such as Davies-Bouldin, Dunn and
Silhouette indices available in
get_internal_clustering_criteria function
- Reordered parameters to
clusters then
true_labels for better intuitive use
- Modified output to return results as a list
- Added parallel execution support for the
generate_indices function
- Implemented kmeans++ initialization for improved clustering
- Added support for additional parameters in
EHyClus,
generate_indices, and funspline
Removals
- Removed the grid parameter (evaluation grid) as it wasn’t affecting
results
- Removed unused parameters for code cleanliness
Documentation Updates
- Updated citations and added inst/CITATION
- Enhanced vignettes
- Added section for automatic variable selection in second
vignette
- Updated examples and documentation