Hugging Face's logo

Datasets:
biglam
/
lampeter_corpus

Tasks:
Text Classification
Sub-tasks:
multi-label-classification
multi-class-classification
Languages:
English
Size:
n<1K
License:
Dataset card Files
xet
 Community
4
Dataset Viewer
The dataset viewer is not available for this dataset.
Cannot get the config names for the dataset.
Error code:   ConfigNamesError
Exception:    RuntimeError
Message:      Dataset scripts are no longer supported, but found lampeter_corpus.py
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/dataset/config_names.py", line 67, in compute_config_names_response
                  config_names = get_dataset_config_names(
                      path=dataset,
                      token=hf_token,
                  )
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 161, in get_dataset_config_names
                  dataset_module = dataset_module_factory(
                      path,
                  ...<4 lines>...
                      **download_kwargs,
                  )
                File "/usr/local/lib/python3.14/site-packages/datasets/load.py", line 1217, in dataset_module_factory
                  raise e1 from None
                File "/usr/local/lib/python3.14/site-packages/datasets/load.py", line 1177, in dataset_module_factory
                  raise RuntimeError(f"Dataset scripts are no longer supported, but found {filename}")
              RuntimeError: Dataset scripts are no longer supported, but found lampeter_corpus.py

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

The Lampeter Corpus of Early Modern English Tracts

120 complete tracts and pamphlets published in England between 1640 and 1740 (earliest dated text 1641, latest 1739), compiled at Chemnitz University of Technology by Josef Schmied, Claudia Claridge and Rainer Siemund with funding from the Deutsche Forschungsgemeinschaft, and distributed by the Oxford Text Archive. The texts come from the Tract Collection of the Founders' Library at the University of Wales, Lampeter, largely a bequest of the Bowdler family.

The corpus is balanced by construction: six subject domains × ten decades from the 1640s to the 1730s × two texts each. All 120 texts have different authors; major literary figures were excluded deliberately, and only first editions were used apart from a few later editions revised by their original authors. Individual tracts run 3,000–20,000 words. They were keyed directly from the 17th- and 18th-century originals rather than from modern editions, so original spelling, punctuation and word division are retained — expect unnormalised orthography before pointing any modern lexicon-based tool at it.

Fields

field notes
id e.g. SciB1735 — domain prefix, sequence letter, year
text body paragraphs, joined (see below)
date year of publication, parsed from the last four characters of id
genre Law, Economy, Religion, Politics, Science or Miscellaneous, derived from the first character of id; 20 texts each
head first heading in the body; empty for 14 of the 120 texts
title short title

The repo still ships a legacy Python loading script, which datasets no longer runs (RuntimeError: Dataset scripts are no longer supported). Until it is converted to Parquet, work from the TEI XML in the OTA record.

What text is not

text is assembled from the <p> elements of the TEI <body> only. The <front> matter (title pages, dedications, prefaces) and <back> matter (postscripts, appendices) are dropped, as is any non-paragraph content inside the body. On the OTA TEI file that leaves roughly 935,000 whitespace-separated tokens against the 1,193,385 words reported for the full corpus. The missing fifth is precisely the paratext the compilers went out of their way to include: the corpus was built to hold complete texts, title page to FINIS. Nine of the 120 texts wrap several sub-texts in a <group>, and only the first sub-text's body is carried across.

The TEI headers also hold Wing STC numbers (on 72 of 120 texts), Lampeter shelfmarks, print place, printer and bookseller names, and page format (on 62). None of that reaches these fields; read the source XML if you need it.

Use

  • Diachronic study of Early Modern English over a decade-balanced sample — date plus text.
  • Register and subject-domain comparison across the six areas.
  • genre classification, bearing in mind that 120 documents at 20 per class is evaluation-sized, not training-sized, and that the perfectly balanced design flatters accuracy figures.

Selection compounds: what one Welsh college library happened to hold, narrowed by a single family's bequest, narrowed again by the compilers' sampling frame. Most tracts were printed in London — a consequence of how printing and press licensing were organised in early modern Britain, not of the topics covered.

Licence

CC BY-SA 3.0, per the Oxford Text Archive record. Share-alike propagates to derivatives.

Citation 

@misc{20.500.12024/3193,
  title = {The Lampeter Corpus of Early Modern English Tracts},
  author = {Schmied, Josef and Claridge, Claudia and Siemund, Rainer},
  url = {http://hdl.handle.net/20.500.12024/3193},
  note = {Oxford Text Archive},
  copyright = {Distributed by the University of Oxford under a Creative Commons Attribution-ShareAlike 3.0 Unported License}
}

Corpus description: Siemund, Rainer and Claudia Claridge (1997). "The Lampeter Corpus of Early Modern English Tracts". ICAME Journal 21, 61–70. Full manual: Claridge, Claudia (1999/2003), "Life is ruled and governed by opinion": The Lampeter Corpus of Early Modern English Tracts. Manual of Information.
Downloads last month
46
Total file size:
10.9 kB

Models trained or fine-tuned on biglam/lampeter_corpus