Run Machine Learning Model

Once you have created a machine learning model you can run the model on other tables. For example, if you created a model using the Zones table, you can apply that model to the Grids table to generate grid-based versus borehole-based predictions. An online document —Production Prediction Workflow—details the steps for creating a production prediction model using the Kingdom Zones table, and also shows how to set optimal engineering constants and grid data to run the model on the Grids table.

The general steps to run the model on another table are the following:

1. In Spotfire, select Analytics Explorer > Machine Learning > Run Machine Learning Model
2. Browse and select the model file. The file will have a .model extension. Remember, the model includes all input attributes, the algorithm, and the hyperparameter settings.
3. Select the table to Apply the Model to. In the Production Prediction Workflow the model created with the Zones table is applied to the Grids table.
4. If you want to limit the analysis to marked rows, check Limit data using markings. The default is to use all data in the selected table.
5. Select the column to predict.
6. Now map each input column in the model to either a column in the table that you are applying the model to, or enter a constant value that you have determined from the machine learning model or other engineering data. Again, see the Production Prediction workflow for a specific example.
7. When all input columns have been mapped, click Run.

The output visualizations will depend on the template and workflow.