"split operator - export data not complete for further use (operators)"

joeijoei MemberPosts:6Contributor I
edited June 2019 inHelp
Hello,

the split operator gives me only the first three columns for further use even if the operator created more. That means that in the result view I see all split columns (more than thee) but I cannot choose them in another operator (only the first three are visible).

Here is a simple table one can try it:
bla split
asdf 2345x2134
dsaf 2345x2345x345x456x356x3546
sadf 2435x2345
Tagged:

Answers

  • Marco_BoeckMarco_Boeck Administrator, Moderator, Employee, Member, University ProfessorPosts:1,984RM Engineering
    Hi,

    my quick test process worked fine, I could select up to "split_6" attribute in further operators:






    <宏/ >

















    <连接from_op =“分裂”from_port设置ou = "例子tput" to_op="Filter Examples" to_port="example set input"/>









    Can you provide your process XML which does not work?

    Regards,
    Marco
  • joeijoei MemberPosts:6Contributor I
    of course. (my post wasn't complete. accidently created two posts...)





    <宏/ >
























    <连接from_op =“分裂”from_port设置ou = "例子tput" to_op="Multiply" to_port="input"/>










  • Marco_BoeckMarco_Boeck Administrator, Moderator, Employee, Member, University ProfessorPosts:1,984RM Engineering
    Hi,

    1. RapidMiner 5.3 is old. Like really old. We cannot provide help for that anymore here. Please consider using RapidMiner Studio 7.0. instead.
    2. You are using the split operator after "Read Excel". The problem is that the output of Read Excel depends on actually reading the excel file at runtime. So until then, we don't know what the result will be. Therefore the split operator creates a dummy output to show an example of how it could look like.
    To use actual data, load it into the repository first, then access it with a "Retrieve" operator. That way, you have full metadata available and the split operator preview will be correct.

    Regards,
    Marco
  • joeijoei MemberPosts:6Contributor I
    The filter example operator also works in my example.
    But I still cant see the split columns higher than 3 in the operators select attributes, rename, remove duplicates (subset).
  • Marco_BoeckMarco_Boeck Administrator, Moderator, Employee, Member, University ProfessorPosts:1,984RM Engineering
    Hi,

    yes, that is expected due to the "can't know beforehand" problem. You can still manually change those parameters if you know you will end up with 6 splits for example.
    But the easiest solution is to read the data into your repository, then only use the data from the repository in your process. That way you have the actual information available during construction time.

    Regards,
    Marco
  • joeijoei MemberPosts:6Contributor I
    ok thank you.
  • joeijoei MemberPosts:6Contributor I
    How does it work with the manually change? The data is to big for loading it into the repository.
  • Marco_BoeckMarco_Boeck Administrator, Moderator, Employee, Member, University ProfessorPosts:1,984RM Engineering
    Hi,

    your local repository sits on your file system - data cannot be to big for that;)
    Manually depends on the parameter. For example for "Remove Duplicates", you can select 'subset', then add the name like "split_6" to the upper right textfield and press +

    Regards,
    Marco
Sign InorRegisterto comment.