README.md

metadata

language:
  - en
license: mit
size_categories:
  - 100K<n<1M
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
      - split: validation
        path: data/validation-*
dataset_info:
  features:
    - name: prompt
      dtype: string
    - name: story
      dtype: string
  splits:
    - name: train
      num_bytes: 883772040
      num_examples: 272600
    - name: test
      num_bytes: 49072859
      num_examples: 15138
    - name: validation
      num_bytes: 50331685
      num_examples: 15620
  download_size: 606854859
  dataset_size: 983176584

Dataset Card for "writingprompts"

WritingPrompts dataset, as used in Hierarchical Neural Story Generation. Parsed from the archive