When trained with very little data, some ensembles may fail to generate due to data shortage. For now the model did not warn the users about it and continues to train all available ensembles. This is not ideal since it indicates that the users have too little data and AdaSTEM is likely unsuitable for their purposes.
Solution: terminate the training when the available ensembles does not match with targeted ensembles.
When trained with very little data, some ensembles may fail to generate due to data shortage. For now the model did not warn the users about it and continues to train all available ensembles. This is not ideal since it indicates that the users have too little data and AdaSTEM is likely unsuitable for their purposes.
Solution: terminate the training when the available ensembles does not match with targeted ensembles.