The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here
Loop Cluster - exclude examples
Contributor I
Hi,
I have used Rapidminer to create many files with clusters. Therefore, my data file has many examples with a set of numeric attributes, a label, and a special attribute - cluster attribute. Next, I am trying to use (Loop Clusters) process to run classification models within validation process which is primarily working for the data files I have. However, some data files have clusters with one or 2 examples or observations; therefore an error is generated whenever the loop reaches any of the clusters with small number of examples. My question is how can I exclude clusters with low number of examples in the Loop Clusters process.
Thanks
I have used Rapidminer to create many files with clusters. Therefore, my data file has many examples with a set of numeric attributes, a label, and a special attribute - cluster attribute. Next, I am trying to use (Loop Clusters) process to run classification models within validation process which is primarily working for the data files I have. However, some data files have clusters with one or 2 examples or observations; therefore an error is generated whenever the loop reaches any of the clusters with small number of examples. My question is how can I exclude clusters with low number of examples in the Loop Clusters process.
Thanks
Tagged:
0
Answers
you can extract the size of your cluster subset with the Extract Macro operator and the option "number of examples", next you use a branch operator, set as condition that the number of examples has to be greater two, and perform your model building in the Then-branch. The Else-branch can return an empty example set or some dummy data, whatever fits best.
Best,
David