README.md

---
dataset_info:
  features:
  - name: messages
    list:
    - name: role
      dtype: string
    - name: content
      dtype: string
  - name: metadata
    struct:
    - name: task_type
      dtype: string
    - name: description
      dtype: string
    - name: room_type
      dtype: string
    - name: difficulty
      dtype: string
    - name: num_steps
      dtype: int64
    - name: has_recovery_steps
      dtype: bool
    - name: has_close_open
      dtype: bool
    - name: has_admissible
      dtype: bool
    - name: num_actions
      dtype: int64
    - name: num_observations
      dtype: int64
    - name: num_detours
      dtype: int64
    - name: num_recovery_steps
      dtype: int64
    - name: has_open_action
      dtype: bool
    - name: subgoals
      sequence: string
    - name: total_chars
      dtype: int64
    - name: trajectory_outcome
      dtype: string
    - name: failure_reason
      dtype: string
    - name: type
      dtype: string
    - name: table_name
      dtype: string
    - name: sql
      dtype: string
    - name: label
      sequence: string
    - name: wrong_sql
      dtype: string
    - name: wrong_val
      dtype: string
    - name: correct_val
      dtype: string
    - name: where_col
      dtype: string
    - name: error_col
      dtype: string
    - name: error_reason
      dtype: string
  - name: id
    dtype: string
  splits:
  - name: train
    num_bytes: 2327351.4671532847
    num_examples: 1197
  download_size: 216617
  dataset_size: 2327351.4671532847
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---