1. Administering Planning
  2. Configuring IPM
  3. Bring Your Own ML: About Machine Learning Model Import
  4. Deploying an ML Model to Planners

Deploying an ML Model to Planners

Deploy an imported ML model to make it available to planners on forms and dashboards. Planners can use machine learning to generate a prediction.

After importing a PMML model, EPM administrators integrate the PMML file into the Planning application by associating the generated Groovy rule with Planning forms.

You can first review the Groovy rules in Calculation Manager to validate the mappings and make changes if needed. When the Groovy rule is finalized, you can associate it with relevant forms or dashboards.

Make the Groovy rule available to planners by creating an Action Menu that you associate with a form.

To associate an ML Groovy rule with a form:

  1. Create an Action Menu item:
    1. From the Navigator, under Create and Manage, click Action Menus.

    2. Click Create Menu icon, enter the menu's name, and then click OK.

    3. Select the new menu item, click Edit Menu icon, and then click Add Child.
    4. Enter a Menu Item name, and give it a label, for example, Predict Volume, of type Business Rule, and select the appropriate cube and Groovy rule with the Form suffix, for example: ML_MLModelName_Form. Then click Save.

    For more information, see Creating and Updating Action Menus.

  2. Associate the action menu with a form:
    1. From the Navigator, under Create and Manage, click Forms.
    2. Navigate to the form, click Edit Menu icon, and under Other Options, click the Action Menu item you created. Associate the menu with the form by moving it from Available Menus to Selected Menus. Then click Save.

    For more information, see Administering Forms.

When a planner selects the Action Menu item from the form, it triggers the Groovy rule. The rule picks the data for all input drivers, sends it to the PMML file for processing, returns the set of prediction values, and pastes them to the output, as defined in the Import Model wizard. The rule runs in the context of the form, predicting values only for cells on the form. Security is honored so that planners see predictions only for intersections to which they have access. Planners can run through various what-if scenarios to adjust their forecasts and plans, or adjust the predicted values.

Note that planners must have access to the model, form, rule, and members in order to run the rule.

Running an ML Groovy Rule as a Batch Job

You can run the ML Groovy rule as a job. The batch rule runs for all cells for the full scope of the model, not just for cells on a particular form. You can set it up to run as a recurring job, for example, to load predictions so they are available to planners on a regular basis.

To run or schedule a batch job, in Jobs, select a Job Type of Rules, and then select the Groovy batch version of the business rule for the PMML model—ML_MLModelName.

For information about scheduling jobs, see Scheduling Jobs.

Tutorials

Tutorials provide instructions with sequenced videos and documentation to help you learn a topic.

Your Goal Learn How
Learn how to take the Groovy rule created from an imported ML Model, and add it to a form. Then you make predictions with the ML model in Planning . tutorial icon Making Predictions with ML Models