Awesome

The-Pile-PhilPapers

Download, parse, and filter an open access collection of philosophy publications PhilPapers, data-ready for The-Pile.

The PhilPapers (PP) are indexed using OAI-MPH, the Open Archives Initiative Protocol for Metadata Harvesting. As such, the first step to collect the data is to get the XML for all links. This was done using a harvester modified from here:

python pyoaiharvest.py -l https://philarchive.org/oai.pl -o data/phil_meta.xml

From that, each publication is downloaded. Some entries do not exist, or have been removed by the authors. Papers with text are extracted using pdfbox, but papers with non-machine readable text are ignored. Non-English language publications are kept, and the metadata reflects the language reported by the OAI-MPH XML. The text is filtered with pdf_filter.py from PDFextract

Pile-V2 Statistics

✔ Saved to PhilArchive.jsonl.zst
ℹ Collection completed 12/15/2021
ℹ 42,464 publications (8,474 added, 24.9% growth)
ℹ Uncompressed filesize 3,270,636,340
ℹ Compressed filesize     985,311,313
ℹ sha256sum 9311a57fcbde8dd832e954821bdf0e1f3e2899d9567f6c3b5d7a2d1161fa3e7d

Pile-V1 Statistics

✔ Saved to PhilArchive.jsonl.zst
ℹ 33,990 publications
ℹ Uncompressed filesize 2,610,566,629
ℹ Compressed filesize     797,708,027
ℹ sha256sum e90529b9b3961328d1e34b60534a8e0f73d5ad1f104e22a217de53cd53c41fea

Stat Sheet created for The-Pile V1.