US Patent 11461648 Standardizing disparate data points

In an approach to deploying a test input based on standardized disparate data points, one or more computer processors perform one or more tests on a test input resulting in one or more disparate data points that are either non-binary or binary. The one or more computer processors standardize the one or more disparate data points utilizing a trained binary classification model. The one or more computer processors generate one or more non-binary machine learning evaluation metrics based on the one or more standardized disparate data points. The one or more computer processors, responsive to the one or more generated non-binary machine learning evaluation metrics exceeding one or more thresholds, deploy the test input.