README.md

metadata

dataset_info:
  features:
    - name: message_id
      dtype: string
    - name: text
      dtype: string
    - name: event_id
      dtype: string
    - name: event_type
      dtype: string
    - name: sub_type
      dtype: string
    - name: urgency
      dtype: string
    - name: language
      dtype: string
    - name: is_noise
      dtype: bool
    - name: source_group
      dtype: string
    - name: timestamp
      dtype: string
    - name: confidence_score
      dtype: float64
    - name: cluster_size
      dtype: int64
    - name: is_contradictory
      dtype: bool
    - name: ground_truth
      dtype: string
    - name: text_length
      dtype: int64
    - name: word_count
      dtype: int64
    - name: __index_level_0__
      dtype: int64
  splits:
    - name: train
      num_bytes: 2446560
      num_examples: 10332
  download_size: 313575
  dataset_size: 2446560
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*