tokenized_news_2 / README.md
falba's picture
Upload dataset
6075065 verified
metadata
dataset_info:
  features:
    - name: document
      dtype: string
    - name: summary
      dtype: string
    - name: input_ids
      sequence: int32
    - name: attention_mask
      sequence: int8
    - name: labels
      sequence: int64
  splits:
    - name: train
      num_bytes: 860380944
      num_examples: 44972
    - name: validation
      num_bytes: 105896194
      num_examples: 5622
    - name: test
      num_bytes: 107677911
      num_examples: 5622
  download_size: 536056058
  dataset_size: 1073955049
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
      - split: test
        path: data/test-*