Group K-fold partitioning a dataset
3 ビュー (過去 30 日間)
古いコメントを表示
The scikit-learn package in Python has a Group K-Fold function that allows you to split the data-set into test/train folds while ensuring the same "group" is not present in different folds. This is useful for example in studies where the same subject/person generates multiple data-points and we want to make sure the samples/data-points belonging to the same subject don't appear in both the training and testing folds.
I was wondering if MATLAB has a way to do this or enable this option in cvpartition function or in some other way. The default options only seem to preserve relative class-sizes.
0 件のコメント
回答 (0 件)
参考
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!