README.md

---
dataset_info:
  features:
  - name: context
    dtype: string
  - name: goal
    dtype: string
  - name: belief
    dtype: string
  - name: action
    dtype: string
  - name: meta_data
    struct:
    - name: orig_index
      dtype: int64
  splits:
  - name: train
    num_bytes: 736455
    num_examples: 500
  - name: eval
    num_bytes: 143159
    num_examples: 100
  download_size: 225972
  dataset_size: 879614
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: eval
    path: data/eval-*
---