Datasets:

Modalities:
Image
Languages:
English
ArXiv:
License:
TableEval / LogicNLG /README.md
katebor's picture
Create README.md
a5a71b5 verified
# LogicNLG subset
**[LogicNLG](https://aclanthology.org/2020.acl-main.708/)** a T2T dataset of open-domain tables fetched from Wikipedia and associated with manually annotated natural language statements that can be logically entailed by the given data.
Images of tables are stored in the <code>.zip</code> folders, while textual formats of tables and all metadata are availabe in <code>.json</code> files.
## Data Fields
- <code> instance_id </code> - (str) unique ID of a given instance;
- <code> table </code> - (List[List[str]]);
- <code> ref </code> - (str) gold statement;
- <code> linked_columns </code> - (List[str]);
- <code> title </code> - (str) table title;- <code> template </code> - (str) ;
- <code> template </code> - (str) statement with masked entities;
- <code> table_id </code> - (str) unique ID of a given table;
- <code> wiki_title </code> - (str) title of the source Wikipedia page;
- <code> wiki_url </code> - (str) URL of the source Wikipedia page;
- <code> table_column_names </code> - (List[str]) table header names;
- <code> table_content_values </code> - List[List[str]])table row values;
- <code> html_content </code> - (str) source HTML page of a table;
- <code> snapshot_timestamp </code> - (str) timestamp of the HTML page version in Wikipedia history;
- <code> filename </code> - (str);
- <code> image_name </code> - (str) table image name;
- <code> table_xml </code> - (str) table in XML format;
- <code> table_html </code> - (str) table in HTML format;
- <code> table_latex </code> - (str) table in LaTeX format.