ConceptAI & ML
Data de-duplication
The process of removing duplicate or near-duplicate images between pre-training datasets and test datasets to prevent data leakage.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Dataset splitting1Overfitting1
Lecturers
1Connections
2The lectures that define it or assume it, and the concepts next to it.