js-free-sites / README.md
crumb's picture
Update README.md
51e177a
metadata
dataset_info:
  features:
    - name: content
      dtype: string
    - name: markdown
      dtype: string
    - name: url
      dtype: string
  splits:
    - name: train
      num_bytes: 82642012
      num_examples: 2067
  download_size: 34862377
  dataset_size: 82642012
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

It is what it says on the tin, 2k websites that don't run javascript (aka: probably really old, and simple!). Filtering needs to be done to take out the defunct sites and error messages. I don't expect this is a hard task.