Options

Append error message

PrenticePrentice Member Posts: 66 Maven
edited August 2019 in Help
Hi,

I want to append one attribute to another one that has already been processed, meaning that it is in the form of a vector space model. The attribute that I want to append is the category of the processed example. 
The two attributes are of the exact same length. However, when I try to append these I get the error message: Example sets are not compatible and the operator expected one or more example sets which are compatible to other ones. This usually means that they must have the same number of attributes and attribute names. But as far as my knowledge goes, they are. Is this because of the vector space model? How can I link my category to my exampleset if this doesn't work? I can't link them before the exampleset gets processed because it will then process the category as well, which I don't want.

Thanks
-Prentice

Tagged:

Best Answers

  • Options
    SGolbertSGolbert RapidMiner Certified Analyst, Member Posts: 344 Unicorn
    Solution Accepted
    Hi Prentice,

    I am a little confused about your description (an example would help), but I think that you are using the Append operator incorrectly: the Append operator puts Example sets which have the same columns together, i.e. it works at the ROW level.

    If you want to add a new attribute you have a couple of options:
    1. Use the Join operator (requires an id column or key on both example sets)
    2. Use the Generate Attributes operator, in case the information required for the new attribute can be derived or extracted from other attributes
    3. Use the Merge operator, requires same number of rows (this could be your case)
    Regards
    Sebastian

Answers

  • Options
    PrenticePrentice Member Posts: 66 Maven
    Hi Sebastian,

    After messing about with Join, I think I got it to work. Now my output is an id, my Categorisation attribute and the word matrix of my attribute which needs to be categorised. To add another question, is this (without the id attribute) all that's necessary for a cross validation and classification algorithm to work? I have already made the categorisation attribute a label.

    Thanks
    -Prentice
  • Options
    PrenticePrentice Member Posts: 66 Maven
    Ah ok, thanks a lot! 
Sign In or Register to comment.