Improve access to previously clustered data

Issue #98 new
Robert Leach created an issue

Once a user has clustered their data, we should provide a means by which they can switch back and forth between the clustered & unclustered versions or even between different versions of the clustered data. Or, even when they open their original data file, they should be given the option to view the clustered version or the unclustered version in some way. Perhaps, TreeView should remember which version they were last viewing so that when they open the same original file, it shows them the clustered version if that's what they were viewing last.

As to how to implement the switch, there could be a menu option under the cluster menu item like this:

Cluster
 - View previously calculated
    - unclustered
    - hierarchical 1 clustered on 10/28/2014
    - hierarchical 2 clustered on 10/28/2014
    - kmeans 1 clustered on 10/28/2014

Whether or not we maintain multiple historical cluster results might be a future option, but at least the ability to switch between clustered & unclustered would prevent people from thinking they have to recluster every time (because they may not realize they must open the cdt file).

We could also/alternatively have a radio button or select list on the right side of the image.

Comments (9)

  1. Christopher Keil repo owner

    The new file system I implemented seems like a good start to implement this, since all clustered files are being numbered and the original file is preserved. We can discuss this feature in the next meeting and open it if we decide to go forward.

  2. Log in to comment