AnDi 2 Benchmark dataset

The benchmark dataset of the 2nd AnDi Challenge has been developed to test and compare methods for analyzing diffusion trajectories with diffusive changes. It explores biological and physical scenarios such as multistate diffusion, trapping, confinement, and dimerization. Each experiment targets specific characteristics—ranging from broad diffusive parameters to extreme trapping conditions—offering a comprehensive foundation for evaluating changepoint detection, classification, and characterization methods. For further details and documentation, see the associated content: Website of the project: http://andi-challenge.org/ Website of the Challenge: https://codalab.lisn.upsaclay.fr/competitions/16618 Tutorial explaining how this dataset has been generated: https://github.com/AnDiChallenge/andi_datasets/blob/master/source_nbs/tutorials/challenge_two_benchmark.ipynb Preprint explaining all details of the 2n AnDi Challenge: https://arxiv.org/abs/2311.18100第二届AnDi挑战赛（AnDi Challenge）基准数据集旨在测试并对比针对存在扩散变化的扩散轨迹的分析方法。该数据集覆盖了多态扩散、捕获、受限运动以及二聚化等生物与物理场景。每项实验均针对特定特性设计，范围涵盖从宽泛的扩散参数区间到极端捕获条件的各类情形，可为变点检测、分类与特征表征方法的评估提供全面的支撑基础。如需了解更多细节与文档，请参阅以下关联内容： 项目官网：http://andi-challenge.org/ 挑战赛官网：https://codalab.lisn.upsaclay.fr/competitions/16618 本数据集生成方式说明教程：https://github.com/AnDiChallenge/andi_datasets/blob/master/source_nbs/tutorials/challenge_two_benchmark.ipynb 第二届AnDi挑战赛全细节预印本：https://arxiv.org/abs/2311.18100