Evaluation tab on a particular predictive query’s detail page to analyze its performance. By default, Kumo use the most recent time window to create your predictive query’s evaluation metrics.
Column analysis is another useful mechanism that Kumo provides diagnosing poorly performing predictive queries. By analyzing these charts, you can better understand how individual values within each column positively or negatively affect the final prediction distribution.
The job details for each of your batch predictions also displays data distribution drift statistics—these metrics are crucial for detecting unexpected changes in the data used to generate your batch predictions.
What types of model diagnostics does Kumo provide?
Suggest Edits
Kumo provides several mechanisms that allow you to validate that your model is providing accurate predictions and providing value to your organisation, as well as verify that your batch predictions are continuously providing accurate predictions. For example, you can click on the