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.
"Read sparse data problem, Is it a bug?"
siamak_want
Member Posts: 98 
Contributor II
Hi,
I use RM 5.2;
I have a sparse dataset in RM Sparse Format. I used the "Read Sparse Operator" for reading my dataset. Because my label is the last token of the dataset, I set the format to "xy". So I have not any ".aml" file and nor label files. I ran the "Read Sparse Operator" and it worked fine and recognizes the "label attribute" very well.
Then I connected a KNN classifier to the "Read Sparse Operator" and I got the error: "Input example set must have special attribute 'label'."!!!
If RM has identified the label, so why it produces such an error when it is being connected to a classifier? now a more surprising fact: The process run with no problem, despite the generated error!!!
Is it a bug in RM or have I done something wrong?
Any help would be appreciated.
I use RM 5.2;
I have a sparse dataset in RM Sparse Format. I used the "Read Sparse Operator" for reading my dataset. Because my label is the last token of the dataset, I set the format to "xy". So I have not any ".aml" file and nor label files. I ran the "Read Sparse Operator" and it worked fine and recognizes the "label attribute" very well.
Then I connected a KNN classifier to the "Read Sparse Operator" and I got the error: "Input example set must have special attribute 'label'."!!!
If RM has identified the label, so why it produces such an error when it is being connected to a classifier? now a more surprising fact: The process run with no problem, despite the generated error!!!
Is it a bug in RM or have I done something wrong?
Any help would be appreciated.
Tagged:
0
Answers
the metadata view is not always accurate - one just cannot predict what will happen sometimes before actually doing it - like loading data from a file. In your case you can either chose to ignore the error because as you found out it's not a real problem, or you can store the data in the repository before proceeding with your k-NN operator, so RM knows your exampleset has a label.
Regards,
Marco
Now I got it, thanks to your nice guide. In fact, before reading the file RM must not be able to find label. I mean RM recognizes thelabel in runtime.
Thanks Marco.