uTHCD是一个全面的非约束性泰米尔手写文字数据库，包含约91000个样本，覆盖156个字符类别，每个类别约600个样本。数据集由阿曼技术大学信息技术系创建，通过邀请志愿者在特定网格内书写样本来收集离线样本，而在线样本则通过数字书写板收集。数据集涵盖了广泛的书写风格和由于离线扫描过程产生的固有畸变，如笔划不连续性和笔划厚度变化等。uTHCD旨在为泰米尔手写文字识别研究提供新的基准，并作为文档图像分

arXiv2021-03-13 更新80

Benchmarking datasets for scMINER clustering analysis

单细胞RNA测序

分类与聚类基准测试

Here are the 13 datasets used in the clustering benchmark of MICA (https://github.com/jyyulab/MICA), the cell clustering engine of scMINER(https://github.com/jyyulab/scMINER). The first 10 datasets (Y

Zenodo2025-03-17 更新60

Dataset for "A fast dictionary-learning-based classification scheme using undercomplete dictionaries"

手写识别

字典学习

This dataset contains a preprocessed version of the publicly available MNIST handwritten digit dataset, formatted for use in the research paper "A fast dictionary-learning-based classification scheme

DataCite Commons2025-06-19 更新100

Comparison of the classification performance by the ResNet-50 and ResNet-101 architectures on the CIFAR-10 and CIFAR-100 datasets.

分类与聚类基准测试

深度神经网络架构对比

Comparison of the classification performance by the ResNet-50 and ResNet-101 architectures on the CIFAR-10 and CIFAR-100 datasets.

NIAID Data Ecosystem70