Sorry to bother. I just transfered from version 4.1 beta2 to the newest version 4.4. But I find I am lost. It seems that there are many changes and improvements. Where can I find all the changes from version 4.1 to version 4.4. I can not find the methods or classes bellowing: IdUtils.getExampleFromId(es, d)); IdUtils.getIdFromExample(e) FlatClusterModel, KMeansClusterModel, CentroidBasedClusterModel... Also I have some problems with wvtoolOperator.setListParameter.
List
I got the error says thatThe method setListParameter(String, List) in the type Operator is not applicable for the arguments (String, List). Can anybody help? Thanks a million.
IngoRMAdministrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University ProfessorPosts:1,751RM Founder
Hi Amy,
I just transfered from version 4.1 beta2 to the newest version 4.4. But I find I am lost.
No wonder. You missed more than one year of development - and we tried to improve a lot...
Where can I find all the changes from version 4.1 to version 4.4.
In each download file / RM installation you will find a change log file called "CHANGES.txt" which describes almost all (at least all important) changes. This is also part of the CVS (in the root of the project).
We completely revised the clustering in the meantime since relying on the IDs was not really reliable and slower than necessary. And I always found working with IDs instead of example and storing the examples (which should actually not be stored at all) less intuitive and not as clean as it could be. As a consequence, the clustering algorithms have changed a lot. Here are some hints:
- The cluster models are now located in "com.rapidminer.operator.clustering" - From there you can get also the ID of how to work without IDs - the IdUtils have been completely removed
I got the error says that The method setListParameter(String, List) in the type Operator is not applicable for the arguments (String, List).
That's quite easy to fix. Just replace the Object[] by a String[], i.e. like in
List textList = new LinkedList(); textList.add(new String[] {"graphics","sample/data/newsgroup/graphics"}); textList.add(new String[] {"hardware","sample/data/newsgroup/hardware"}); wvtoolOperator.setListParameter("texts", textList);
Hope that helps. By the way: I moved this topic into the board "Development". Ingo
Hi Ingo, Thanks a million for your patience and support. I sincerely appreciate all the replies. It really helps a lot. I will try to catch up. Thanks.
Answers
In each download file / RM installation you will find a change log file called "CHANGES.txt" which describes almost all (at least all important) changes. This is also part of the CVS (in the root of the project). We completely revised the clustering in the meantime since relying on the IDs was not really reliable and slower than necessary. And I always found working with IDs instead of example and storing the examples (which should actually not be stored at all) less intuitive and not as clean as it could be. As a consequence, the clustering algorithms have changed a lot. Here are some hints:
- The cluster models are now located in "com.rapidminer.operator.clustering"
- From there you can get also the ID of how to work without IDs - the IdUtils have been completely removed
That's quite easy to fix. Just replace the Object[] by a String[], i.e. like in
Hope that helps. By the way: I moved this topic into the board "Development".
Ingo
Thanks a million for your patience and support.
I sincerely appreciate all the replies. It really helps a lot.
I will try to catch up. Thanks.
Best Regards
Amy