Importance of Disjoint Sampling in Conventional and Transformer Models for Hyperspectral Image Classification

arXiv:2404.14944v1 Announce Type: new
Abstract: Disjoint sampling is critical for rigorous and unbiased evaluation of state-of-the-art (SOTA) models. When training, validation, and test sets overlap or share data, it introduces a bias that inflates performance metrics and prevents accurate assessment of a model's true ability to generalize to new examples. This paper presents an innovative disjoint sampling approach for training SOTA models on Hyperspectral image classification (HSIC) tasks. By separating training, validation, and test data without overlap, the proposed method …

abstract art arxiv assessment bias classification cs.cv data eess.iv evaluation image importance metrics performance sampling sota state test training transformer transformer models true type unbiased validation