Concept
OOD
OOD stands for out-of-distribution: test-time inputs drawn from a different distribution than the training data. It is the practical opposite of the IID assumption most models are trained under, where train and test samples come from the same source.
The rest of “OOD” 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→