Due to recent updates, all users are required to create an Altair One account to login to the RapidMiner community. Click the Register button to create your account using the same email that you have previously used to login to the RapidMiner community. This will ensure that any previously created content will be synced to your Altair One account. Once you login, you will be asked to provide a username that identifies you to other Community users. Email us at Community with questions.
Normalization Issue and RapidMiner Community 4.2 (Windows 32)
I cannot get RapidMiner Community 4.2 (Windows 32) to properly normalize output variables. I've created identical Operator Trees in versions 4.1 and 4.2, and while version 4.1 will properly normalize output values, version 4.2 will not. To test, create a simple linear regression problem then save the output model. Then create another tree to import the saved model and data--when the program is run the results will not be normalized. Has anyone else had this same problem in v 4.2? Thanks...
0
Answers
unfortunately I do not quite understand your problem. What exactly is not normalized? It would be great, if you could explain it a little bit more in depth. And would you mind sending us the process you consider as to give faulty results?
Thanks!
Regards,
Tobias
I think I have a clue which makes the difference between 4.1 and 4.2: did you activate the "create_preprocessing_model" parameter for the Normalization operator? In versions < 4.2 you did not have to combine the models (the preprocessing model and the prediction model) yourself but this was done automatically by RapidMiner. Unfortunately, this automatic combination was rather restrictive for more complex models and so we replaced the automatic combinations with two operators "GroupModel" and "UngroupModel" (see the change log for version 4.2) which can be used to group two or more models (like the grouping in powerpoint etc.). This gives you more flexibility in creating and, well, grouping your models.
If you have a setup like
- Load Data
- Normalize Data and create preprocessing model
- Learn Prediction Model
- Save (automatically combined) Model
you can simply add the model grouping operator like in this example:
- Load Data
- Normalize Data and create preprocessing model
- Learn Prediction Model
- apply "GroupModel" <-- new step
- Save (automatically combined) Model
Hope that helps,
Ingo