Export Example set and Performance Vector

Anna_May1
Anna_May1 New Altair Community Member
edited November 5 in Community Q&A
Hi guys, 

as part of my sentiment analysis I am receiving results in the "Results" tab, which are named "PerformanceVector (Performance)" and "ExampleSet (Apply Model(2)).
I want to export them into an Excel file exactly as shown in the two pictures below, so exactly the way they are in the Results tab.

But when I e.g. use the Write Excel operator behind apply model, the columns "prediction" both "confidence" columns and the "text" column are gone and I need them in my exported file. The results are then like this: 


Have you guys got any solution for that issue?

Cheers!

Best Answer

  • MartinLiebig
    MartinLiebig
    Altair Employee
    Answer ✓
    Hi,
    are you sure that the two columns aren't by accident on the far right of your dataset?

    Best,
    Martin

Answers

  • MartinLiebig
    MartinLiebig
    Altair Employee
    Answer ✓
    Hi,
    are you sure that the two columns aren't by accident on the far right of your dataset?

    Best,
    Martin
  • Anna_May1
    Anna_May1 New Altair Community Member
    Hi @mschmitz,

    you're totally right! It was there! Thanks a lot!

    Still, when extracting "PerformanceVector (Performance)" in Excel, there are not all the results shown:

    Do you have any idea why that might be the case?
    I have used the "Confusion Matrix to ExampleSet" Operator in front of the "Write Excel" Operator.

    Cheers!
  • MartinLiebig
    MartinLiebig
    Altair Employee
    Hi,
    likely you want to use also performance to data, to convert the performance measures so that you can then write them away?
    Best,
    Martin
  • Anna_May1
    Anna_May1 New Altair Community Member
    Hi @mschmitz

    I tried that, once with"Perfomance to data" on its ows before "Write Excel" and once in combination with "Confusion Matrix to ExampleSet". This was the Excel Output: 

    Cheers! 
  • MartinLiebig
    MartinLiebig
    Altair Employee
    Hi,
    so what is wrong with this?

    Best,
    Martin