Skip to content

Latest commit

 

History

History
9 lines (9 loc) · 635 Bytes

File metadata and controls

9 lines (9 loc) · 635 Bytes
  1. New contrived 3 cluster dataset on 'make deployrun' with non-diagonal covariances.
  2. Allow user to choose between diagonal vs full GMM clustering.
  3. Introduce K-means clustering.
  4. Allow user to pre-initialize cluster centers using a. random selection of c data points or means of groups. b. using K-means algorithm.
  5. Ability to generate a chart from a dataset and labels independent from the GMMCluster formula or Clustering dialog.
  6. Improve notebook bar integration. (Currently no icon for ClusterRows?)
  7. Test GMM using more contrived datasets - centers far away from origin, out of scale dimensions.