Due to recent updates, all users are required to create an Altair One account to login to the RapidMiner community. Click the Register button to create your account using the same email that you have previously used to login to the RapidMiner community. This will ensure that any previously created content will be synced to your Altair One account. Once you login, you will be asked to provide a username that identifies you to other Community users. Email us at Community with questions.

Feature weight operators

ThiruThiru Member Posts: 100 Guru
in Help
I have gone through the recent white paper on model explainability.

I couldnt find the feature weight calculating operators like:  Shap, Shapley.  I am using education studio version 9.9.003
can you guide me how to find them. thanks.

thiru

Best Answers

  • ThiruThiru Member Posts: 100 Guru
    Solution Accepted
    Hi , thanks for your reply.  I m able to see three operators in interpretations extension:  1.  Generate interpretations, 2. Generate neighbourhood,  3.  Generate univariate interpretations.   I think this belongs to 'generate interpretations'.

    thanks.

    thiru

     
  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,527 RM Data Scientist
    Solution Accepted
    exactly. It covers Shaply, KernelShap and Lime.
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany

Answers

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,527 RM Data Scientist
    Hi,
    please check the Interpretations extension.
    Cheers,
    Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • sherlocksherlock Member Posts: 24 Contributor II
    Hi,

    the interpretations extension is working fine with GBM but not with XGBoost (Type / Class issue, it expects AbstractModel). Are there plans in the nearby future to support XGBoost as well? It would allow us to interpret the shortlisted models using the same component / methods (Lime, KernelSHAP, Shapley). The "Explain Predictions" extension works fine and is useful, yet a different algorithm. The results wouldn't be 1:1 compareable.

    Cheers
    Erwin


Sign In or Register to comment.