---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: conversation
    list:
    - name: from
      dtype: string
    - name: speech
      dtype: string
    - name: text
      dtype: string
    - name: unit
      dtype: string
  splits:
  - name: train
    num_bytes: 475340.9755
    num_examples: 100
  download_size: 247344
  dataset_size: 475340.9755
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---