README.md

metadata

license: apache-2.0
dataset_info:
  features:
    - name: xml_sha1
      dtype: string
    - name: declaration_xml
      dtype: string
    - name: declaration_json
      dtype: string
    - name: extracted_text
      dtype: string
  splits:
    - name: train
      num_bytes: 193695685
      num_examples: 10944
  download_size: 35700267
  dataset_size: 193695685
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

We removed all B64 encoded values as they clutter the context without providing meaningful information.