Datasets:

Modalities:
Text
Formats:
json
ArXiv:
Libraries:
Datasets
pandas
License:
File size: 448 Bytes
df820f9
 
 
d602a28
 
c332a94
 
d602a28
1
2
3
4
5
6
7
8
---
license: apache-2.0
---
467 Project Gutenberg books, mostly of older provenance (author died pre-1914), chapterized by Chapter Captor (https://arxiv.org/abs/2011.04163), then requiring the number of chapters is correct and the assigned numbers are sequential starting at 1.

Each line is one chapter of a book.

Keys are "chapter_number", "text", "title", and "metadata{"id"}""  The id is the Gutenberg book number.  Title is often not present.