Pivot Operator - "potential problem detected: attribute missing" after Select Attributes

Ina_KIna_K MemberPosts:9Contributor II
edited November 2018 inHelp

Hi all,

according to the pivot operator in my data preparation process RapidMiner is experiencing a 'potential problem':

18-01-2017_pivot_prob.png

I have to stream a big amount of data.

In Select Attributes I chose three attributes. One of them named TBLUNIQUELRU_ID is being missed by the pivot operator although it is contained in the Select Attributes output data:

Data:SimpleExampleSet: 10000000 examples, 3 regular attributes, no special attributes

A breakpoint is set after the second operator and I can confirm, that the attribute is contained in the pivotinputexample set.

Code:




<参数键= "选项卡le_name" value="ZZ_RM_TEST"/>


















Can someone help?

Ina

Tagged:

Best Answer

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University ProfessorPosts:3,381RM Data Scientist
    Solution Accepted

    Hi Ina,

    propably this is just an issue with the meta data propagation. Isn't there a button to just let it run anyway?


    Otherwise I would recommend to switch the metadata propagation to real data by using Process->Synchronize Data with Real Data.

    Best,

    Martin

    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany

Answers

  • Ina_KIna_K MemberPosts:9Contributor II

    Hello,

    我遇到一个问题属性识别with one of the operators. The example set needs to be streamed, since the amount of data is very big (> 30 mio. examples with corporate license and RM Server).

    To test and work on the process locally I used a small subset of 10000 rows with the Read Database operator.

    每当我使用Read Databasewith a subset of 10000 examples -everything is fine.

    Whenever I incorporateStream DatabaseI encounter a 'Potential problem detected' The Pivot operator doesnt recognize one of the crucial fields (an ID-Field which identifies the examples).

    25-01-2017_read_db_ok.png25-01-2017_read_db_nok.png

    The code with Stream DB:







    <参数键= "选项卡le_name" value="ZZ_RM_TEST"/>







































    I synchronized the meta data with real data as you suggested recently.. Unfortunately it didnt help.

    I am trying to avoid to let it run anyway is because I need to know if this is the reason why the processing takes so long. The process loads up to 16 hours, not coming to an end and getting stuck in the Pivot operator. I really would like to know why this potential problem notification appears and how to solve it. Because now it seems that not only the TBLUNIQUELRU_ID is missing in the input example but the attribute BITID as well.

    25-01-2017_attributes_missing.png

    Or could you explain to me what the issue with the meta data propagation is about?

    Advices are really appreciated.

    Kind regards!

Sign InorRegisterto comment.