Concept
Held-out split
A held-out split is a portion of your data that you deliberately exclude from the fitting process so you can measure performance on examples the optimizer never got to touch. If you already know the classic train/validation/test division, this is the same idea: the held-out…
The rest of “Held-out split” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.
Log in to unlock→