Auto Numeric Visualizations
The following tables and options are available for Auto Numeric visualizations.
Contains a set of one or more models generated by the Auto Numeric node, each of which can be individually viewed or selected for use in scoring. For each model, the model type, the time it took to build the model, the number of fields or features used in the model, the prediction accuracy, and a thumbnail scatter chart of observed and predicted values. The Actions column lets you see the details of an individual model, or remove it from the screen (this does not change the results in the Ensemble Model Evaluation panel).
Ensemble Model Evaluation panel
This panel displays a bar graph showing the overall prediction accuracy, assessed as R2 or squared correlation between observed and predicted valued, and a table containing a set of evaluation statistics (if the prediction accuracy is exactly 0, the graph will not be shown). The evaluation statistics include R2 and may also include additional measures.
Note that all evaluation measures are based on the full ensemble incorporating all models run in the Auto Numeric node. The results do not change if one or models are removed from the ensemble in the output node.
Like your visualization? Why not deploy it? For more information, see Deploy a model.