Important: Submission policy
Parameter tuning is only allowed on the training data. Evaluating on the test data via this evaluation server must only be done for the final system. It is not permitted to use it to train systems on the test data, for example by trying out different parameter values and choosing the best. This is to avoid overfitting on the test data. To help enforcing this policy, we block updates to the test set results of a method for one week after a test set submission. Note that this block will become effective for all test datasets even if only submitting a result for a single test dataset.
Results of different parameter settings of an algorithm can therefore only be reported on the training set. You can split up the training data into training and validation sets yourself as you wish.
It is not permitted to create multiple test set submissions for the same method, or to register on this webpage with multiple e-mail addresses. We will ban users or domains if required.