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

Integrating a trained model

SguishSguish Member Posts: 13 Contributor II
edited November 2018 in Help
Hello, in the last few day I tried integrating a process into a java application. I followed the instructions posted in the FAQ and it seems to work neatly. But then I wondered: what should I do in order to integrate the already trained model? Because trainign could be very lenghty, and of course an application couldn't wait three days each time for the training.

Now, suppose I have a process made of:

1) Input-processing pipeline (replace missing values, selecting attributes etc);
2) Input splitting;
3) Training;
4) Testing.

I'd like to still have 1), but then new data should be fed to a trained model. Maybe I missed something, but this part still puzzles me. What shall I do?
Thanks in advance and good easter, btw!
Tagged:

Answers

  • JEdwardJEdward RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 578 Unicorn
    If you use Store Model in the GUI you can later use that model in your process. 
    Simply Read it from the repositroy & apply.  When calling from code it's exactly the same as it's from a process you have already built.
  • SguishSguish Member Posts: 13 Contributor II
    Great, this is exactly what I needed, thank you!
    One more question then. Now, suppose I integrate the data-processing pipeline and the trained model. Now let's suppose every time new records are added to the database, they must be evauled by the model. Is it correct to add a query component and change the query parameter every time, to select the right ones?
Sign In or Register to comment.