The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here

can someone help me with the data processing ?

tranductrungtranductrung Member Posts: 7 Learner I
I need to run, analyze this data to make predictions through deep learning specifically LSTM layer

Best Answers

  • ruhailaruhaila Member Posts: 48 Guru
    Solution Accepted
    Perhaps share your .rpm file and state what you want to do?
  • ruhailaruhaila Member Posts: 48 Guru
    Solution Accepted
    By doing it yourself, only then you will learn. This community is to help learning.
  • ruhailaruhaila Member Posts: 48 Guru
    Solution Accepted
    When you split train and test data, your FILTER EXAMPLE RANGE is empty because you used 1 - 249 which is the size of your dataset. Lower this number to half maybe. You can use the BREAKPOINT AFTER to see the result of the FILTER EXAMPLE RANGE. hth
  • ruhailaruhaila Member Posts: 48 Guru
    Solution Accepted
    And you also need to set the ID and the LABEL.
  • ruhailaruhaila Member Posts: 48 Guru
    Solution Accepted
    Using process2.rmp, do these:
    1. Use GENERATE ID operator
    2. Use SET ROLE operator
     - set LABEL on the attribute you want to predict e.g. CLOSE
    3. Change the FILTER EXAMPLE RANGE to half
  • ruhailaruhaila Member Posts: 48 Guru
    Solution Accepted
    Hi, 

    Use your file named process2.rmp and do these:
    1. Add GENERATE ID operator
    2. Add SET ROLE operator



     - in the parameters, set the CLOSE attribute with the role LABEL (if CLOSE is what you want to predict)




    3. Change the FILTER EXAMPLE RANGE to half


    hth


  • ruhailaruhaila Member Posts: 48 Guru
    Solution Accepted
    How many neurons depends on what you want to do. You can refer to other works that are similar to your subject.

    hth
  • ruhailaruhaila Member Posts: 48 Guru
    Solution Accepted
    Honestly, I am not familiar with the subject. Perhaps try different runs of 5, 10, 20 neurons? I am just guessing here. hth

Answers

Sign In or Register to comment.