GitHub

该数据集是针对合成GitHub数据集的测试分割部分，其中未对使用的词汇量进行限制。此外，收集的知识图谱来源于GitHub上大于5KB且具有特定扩展名的文件。该数据集规模包含3,922个文件，其任务是对语义表进行解读。

This dataset constitutes the test split of a synthetic GitHub dataset, with no restrictions imposed on the vocabulary utilized. Furthermore, the collected knowledge graphs are sourced from GitHub files that are larger than 5 KB and have specific file extensions. Comprising 3,922 files in total, this dataset is tailored for the task of semantic table interpretation.