Leveraging Green Infrastructure for Efficient Treatment of Reclaimed Water
收藏PAN-X
该数据集是Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME)基准的一部分,名为WikiANN或PAN-X。它包含多种语言的维基百科文章,特别是瑞士四种最常用语言:德语、法语、意大利语和英语。每篇文章都使用LOC(位置)、PER(人物)和ORG(组织)标签在‘inside-outside-beginning’(IOB2)格式下进行了标注。
github 收录
The sex [Male (1) and female (2)] age (in years), weight (in lbs), #GPS Points (total after filtering), and the 100% MCP, 95% KDE, and 50% KDE home ranges (ha) for all cats sampled in the study. All cats were desexed. The personality scores (shown as a percent), were obtained from a survey, based on the “Feline Five” (Litchfield et al., 2017), that evaluated how much owners agreed or disagreed that their cats showed certain traits. Traits were then summed and converted into percentages. Bold cats are considered to have a low neuroticism score. Road density was estimated by summing the road lengths, measured in meters, within a fixed boundary centred on each cat’s mean latitude and longitude coordinates. The variable “major road” indicated the presence (1) or absence (0) of a major road near the cat’s home range. Roads were labeled as “major” based on Google Maps’ classification, related to traffic rates, and through “ground-truthing”.
Domestic cats (<i>Felis catus</i>) play a dual role in society as both companion animals and predators. When provided with unsupervised outdoor access, cats can negatively impact native wildlife and create public health and animal welfare challenges. The effective implementation of management strategies, such as buffer zones or curfews, requires an understanding of home range size, the factors that influence their movement, and the types of habitats they use. Here, we used a community/citizen scientist approach to collect movement and habitat use data using GPS collars on owned outdoor cats in the Kitchener-Waterloo-Cambridge-Guelph region, southwestern Ontario, Canada.
DataCite Commons 收录
学生课堂行为数据集 (SCB-dataset3)
学生课堂行为数据集(SCB-dataset3)由成都东软学院创建,包含5686张图像和45578个标签,重点关注六种行为:举手、阅读、写作、使用手机、低头和趴桌。数据集覆盖从幼儿园到大学的不同场景,通过YOLOv5、YOLOv7和YOLOv8算法评估,平均精度达到80.3%。该数据集旨在为学生行为检测研究提供坚实基础,解决教育领域中学生行为数据集的缺乏问题。
arXiv 收录
mmlu_eval
该数据集用于评估和比较不同模型的推理能力。它包含多个特征,如问题、主题、选项、答案、输入、基线模型输出、混合推理模型输出和评估结果。数据集分为一个验证集,包含1531个样本。数据集的大小为10295402字节,下载大小为4908248字节。
huggingface 收录
AIS数据集
该研究使用了多个公开的AIS数据集,这些数据集经过过滤、清理和统计分析。数据集涵盖了多种类型的船舶,并提供了关于船舶位置、速度和航向的关键信息。数据集包括来自19,185艘船舶的AIS消息,总计约6.4亿条记录。
github 收录