Held-out dataset
收藏Zenodo2026-06-16 更新2026-06-18 收录
下载链接：
https://zenodo.org/doi/10.5281/zenodo.20722590
下载链接
链接失效反馈官方服务：
资源简介：
This is a held-out dataset
应用场景：
提供机构：
Zenodo创建时间：
2026-06-16
相关数据集
Evaluation on the held-out dataset.
The model cross-validated on the ABIDE dataset performs with AUC = 0.707 and ACC = 76% on DS030. The recall column shows the insensitivity of the classifier to the true “exclude” cases. The predicted
Figshare2017-09-25 更新10
Evaluation on the held-out dataset.
The model cross-validated on the ABIDE dataset performs with AUC = 0.707 and ACC = 76% on DS030. The recall column shows the insensitivity of the classifier to the true “exclude” cases. The predicted
NIAID Data Ecosystem30
Predictive held-out likelihood.
Topic models are widely used to discover the latent representation of a set of documents. The two canonical models are latent Dirichlet allocation, and Gaussian latent Dirichlet allocation, where the
Figshare2023-07-12 更新00
Synthetic Held-out Data
该数据集是为了支持数据推断（DI）任务而生成的人工合成数据，其目的是尽可能贴近原始数据集的分布特征。为此，该数据集采用了包括上下文学习、偏好优化和后缀补全在内的多种方法来生成，以最小化与原始数据集分布上的偏差。其规模根据原始数据集的大小而有所不同，所针对的任务是数据推断（DI）。
arXiv150
Predictive held-out likelihood.
Topic models are widely used to discover the latent representation of a set of documents. The two canonical models are latent Dirichlet allocation, and Gaussian latent Dirichlet allocation, where the
NIAID Data Ecosystem00
