The HIPE-eval organisation contains code and datasets related to the HIPE evaluation campaigns on the processing of named entity in historical documents. HIPE-eval is gradually evolving towards hosting other data and developments related the evaluation of information extraction systems on historical documents. HIPE stands for 'Identifying Historical People, Places and other Entities'.
What are the main HIPE resources?:
- HIPE-scorer: a python module for evaluating NERC and NEL systems (CLI-based, soon available via hugging-face).
- HIPE-pycommons: to manipulate data and get some statistics.
- HIPE-2022-data: data of the HIPE 2022 shared task.
- HIPE-2022-eval: all you need to reproduce the HIPE-2022 evaluation campaign results.
- HIPE-2022-baseline: baseline models for NERC for HIPE-2022.
- CLEF-HIPE-2020-eval: all you need to reproduce the HIPE-2020 evaluation campaign results (in impresso GH organisation).
Where can I found more on the evaluation campaigns and their results? 📚
- visit the HIPE-2022 website.
- or the CLEF-HIPE-2020 website.
- and the HIPE Zenodo organisation.
How can I contribute?
You are welcome to use the scorer and the data (please note the licences and refer to the relevant scientific papers). If you find any errors, please feel free to open issues or pull requests.
What's to come?
We hope to be able to offer another edition of HIPE, and to publish useful data again, so stay tuned!