Customizable machine learning models

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for customizable machine learning models. In some implementations, data is received, including (i) example data sets and (ii) data specifying one or more criteria to be assessed. Models are generated based on different combinations of features using training data sets comprising subsets of the example data sets. Output is obtained from the generated models, and one of the combinations of features is selected based on the outputs. The example data sets are used to train a classifier to evaluate input data with respect to the specified one or more criteria based on input values corresponding to the features in the selected combination of features.