currently, if a problem is specified where all the treated/control units have identical covariates, the code throws an error and fails. This should probably be switched over to something that discards that odd set, maybe throws out a warning, but continues running.
currently, if a problem is specified where all the treated/control units have identical covariates, the code throws an error and fails. This should probably be switched over to something that discards that odd set, maybe throws out a warning, but continues running.