Strive for balance between training, tuning, and testing datasets

[hugoaerts]

Scientists should strive for balance in the training, tuning, and testing datasets to assure that different phenotypic groups are represented appropriately/similarly. This refers to the balanced proportion of different classes of outcome or target variables. If class imbalance is inevitable, appropriate strategies like augmentation or bootstrapping can be used.

[Benjamin-Lee]

With augmentation, one needs to be careful to augment the training, testing, and validation data separately, not together (and then splitting) to prevent overfitting.