These settings will have no effect whatsoever in BIRCHCluster. I just
looked at the code: the options for setting boolean and nominal
attributes are inherited from the abstract ClusterGenerator class but
not used in BIRCHCluster at all.
On 20/03/2014 15:14, Navati Jain wrote:
I am using Weka for my research and want to generate synthetic data
which has boolean and nominal data in addition to numeric data using
the BIRCH Cluster (available in clusterers in data generators) in the
Weka Explorer (GUI). There is an option in the BIRCH cluster to have
boolean(or nominal) columns which requires a range to be specified
(booleanCols -- The range of attributes that are generated as boolean
ones). I have tried to work with different ranges but always get the
error message ' Error generating data: premature end of file, read
I would like to understand how to enter the range in this option, I have
looked up a lot on this issue and tried different methods but have been
unable to resolve this problem.
I hope to hear from you soon. Thank you so much.
Department of Computer Science
Wekalist mailing list
Send posts to: Wekalist(a)list.waikato.ac.nz
List info and subscription status: http://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html