metadata

dataset_info:
  features:
    - name: relative_path
      dtype: string
    - name: section
      dtype: string
    - name: filename
      dtype: string
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 304498096
      num_examples: 4754
  download_size: 110063826
  dataset_size: 304498096
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: odc-by
size_categories:
  - 1K<n<10K
task_categories:
  - text-generation
  - feature-extraction

BEE-spoke-data/napierone-pdf-raw

NapierOne PDF files converted with marker. Seems to contain mostly books from Project Gutenberg.

Datasets:

BEE-spoke-data
/

napierone-pdf-raw

BEE-spoke-data/napierone-pdf-raw

detected languages