Currently, when duplicates are encountered in the cohort or label table, this causes a downstream error in trying to save predictions/evaluations (because a duplicate primary key is encountered). We should check for this earlier (as soon as these tables are created) and error immediately rather than proceeding to build models.
Currently, when duplicates are encountered in the cohort or label table, this causes a downstream error in trying to save predictions/evaluations (because a duplicate primary key is encountered). We should check for this earlier (as soon as these tables are created) and error immediately rather than proceeding to build models.