five

pfnet/amb-hmeg

收藏
Hugging Face2026-04-25 更新2026-05-10 收录
下载链接:
https://hf-mirror.com/datasets/pfnet/amb-hmeg
下载链接
链接失效反馈
官方服务:
资源简介:
AmbHMEG(模糊手写数学表达式数据集)是一个由AmbHMEG模型生成的模糊手写数学表达式数据集,用于将LaTeX表达式渲染为视觉上模糊的数学图像。数据集包含三个子集:basic(标准单条件生成数据集)、layout(双条件数据集,包括原始表达式和去除上标与下标的表达式)以及symbol(双条件数据集,包括原始表达式和用视觉相似符号替换的表达式,替换规则涵盖数字、字母、希腊字母和数学符号的混淆对)。数据格式为每个样本包括一个图像(PNG格式)和一个对应的标签文件(可JSON序列化的PKL格式)。数据集基于MathWriting数据集衍生,遵循CC BY-NC-SA 4.0许可,禁止商业使用,衍生作品需共享相同许可并注明原作者。

AmbHMEG (Ambiguous Handwritten Mathematical Expression Dataset) is a dataset of ambiguous handwritten mathematical expressions generated by the AmbHMEG model, which renders LaTeX expressions as visually ambiguous math images. The dataset includes three subsets: basic (standard single-conditioned generated dataset), layout (dual-conditioned dataset with original expressions and expressions with superscripts and subscripts removed), and symbol (dual-conditioned dataset with original expressions and expressions where symbols are replaced with visually similar alternatives, covering confusion pairs for digits, letters, Greek letters, and mathematical symbols). Each sample consists of an image (PNG format) and a corresponding label file (JSON-serializable PKL format). The dataset is derived from the MathWriting dataset and is licensed under CC BY-NC-SA 4.0, prohibiting commercial use and requiring derivative works to be shared under the same license with attribution to the original authors.
提供机构:
pfnet
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作