An evaluation provides an easy way to measure the performance of a predictive model. To create a new evaluation you need a model/id or an ensemble/id and a dataset/id. The complete and updated reference with all available parameters is in our
documentation website.
