"Saving individual training/test sets used in x-validation"

invictus_champ

Crossvalidation automatically splits data into several subsets. Is there a store these internal datasets to individual files at each round?

Thanks.

earmijo

I don't think you can do that. The best alternative I can think of is Batch X-Validation. With it, you define the batches yourself. You have full control over which observations go into each fold.

Andrew2

Hello

The process here http://rapid-i.com/rapidforum/index.php/topic,6629.msg23271.html#msg23271 remembers each training and test fold for later use. You could modify it to store into a set of files.

Andrew