README.md

---
dataset_info:
  features:
  - name: interaction_id
    dtype: string
  - name: turn_id
    dtype: string
  - name: duration
    dtype: float32
  - name: speaker_id
    dtype: string
  - name: content
    dtype: string
  - name: features
    sequence:
      sequence: float32
  splits:
  - name: valid
    num_bytes: 26893045
    num_examples: 28
  - name: test
    num_bytes: 438891375
    num_examples: 457
  download_size: 117280737
  dataset_size: 465784420
configs:
- config_name: default
  data_files:
  - split: valid
    path: data/valid-*
  - split: test
    path: data/test-*
---