NuminaMath-LEAN-Expert-Iteration
收藏资源简介:
该数据集包含491个训练样本,每个样本由四个字段组成:唯一标识符(uuid)、形式化陈述(formal_statement)、非形式化陈述(informal_statement)以及非形式化证明(informal_proof)。数据集总大小为226023字节,下载大小为130683字节。数据以单一训练集(train)形式组织,未提供验证或测试集划分。从字段命名推断,该数据集可能用于形式化数学与非形式化数学表述之间的转换或证明生成任务,但README未明确说明具体应用场景。
This dataset comprises 491 training samples, each containing four fields: unique identifier (uuid), formal_statement, informal_statement, and informal_proof. The total size of the dataset is 226,023 bytes, while its download size is 130,683 bytes. The data is organized as a single training split (train), with no validation or test splits provided. Based on the naming of the fields, it can be inferred that this dataset may be used for tasks such as conversion between formal and informal mathematical expressions or proof generation; however, the README does not explicitly specify the specific application scenarios.
数据集概述
基本信息
- 数据集名称: NuminaMath-LEAN-Expert-Iteration
- 存储库地址: https://huggingface.co/datasets/ChristianZ97/NuminaMath-LEAN-Expert-Iteration
- 下载大小: 130,683 字节
- 数据集大小: 226,023 字节
数据内容与结构
特征字段
- uuid: 唯一标识符,字符串类型。
- formal_statement: 形式化陈述,字符串类型。
- informal_statement: 非形式化陈述,字符串类型。
- informal_proof: 非形式化证明,字符串类型。
数据划分
- 训练集 (train): 包含 491 个样本,大小为 226,023 字节。
配置信息
- 默认配置: 数据文件路径为
data/train-*,对应训练集划分。




