Export trained solutions to production
-
- UpdatedJan 30, 2025
- 2 minutes to read
- Yokohama
- AI Experiences
Refine and test your ML solutions iteratively on a non-production instance, and then use update sets to export the changes to your production instance. This practice mitigates the risk of retraining solutions on your live production instance.
Before you begin
Prior to testing on a test instance, ensure that the instance hosts recently-cloned data from your production instance so that the solutions you train on the test instance remain valid when you export them to production.
Role required: admin or ml_admin
About this task
Plan your changes carefully, and remember that update sets match records based on the system ID [sys-id] and not the version number. Note that moving solutions to instances can sometimes deliver unpredictable results. If you run into such an issue, retrain the solution again, which takes 5 minutes per solution. For more information on update sets, see System update sets .
Procedure
- Navigate to or .
- Click the name of your trained ML Solution Definition record to open it.
- In the Related Links section, click Add solutions to the current update set.
- Click Update.
Result
What to do next