Skip to content

huridocs/pdf-labeled-data

Repository files navigation

PDF labeled data

Labeled data for creating machine learning models related to PDF consumption: token types, paragraph extraction, and reading order

Dependencies

Quick Start

Start the labeling tool:

make start

When ready, check out the web here:

 http://localhost:8080

To stop the server:

make stop

Labeled data

  1. Token Type: Labels each word that appears in a PDF. Check out this repository https://github.com/huridocs/pdf-tokens-type-labeler

  2. Reading Order: Sorts the information in a PDF https://github.com/huridocs/pdf-reading-order

  3. Paragraph Extraction: Segments a PDF in paragraphs https://github.com/huridocs/pdf_paragraphs_extraction

  4. Table Of Content: Extracts the Table Of Content https://github.com/huridocs/pdf_paragraphs_extraction

About

This is a fork, supported by HURIDOCS, of the Allen AI project PAWLS https://github.com/allenai/pawls

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published