README.md

---
dataset_info:
  features:
  - name: instruction
    dtype: string
  - name: input
    list:
    - name: class
      dtype: string
    - name: object_id
      dtype: int64
    - name: situation
      dtype: string
    - name: subclass
      dtype: string
  - name: output
    dtype: string
  - name: class
    dtype: string
  - name: img_path
    dtype: string
  - name: language
    dtype: string
  splits:
  - name: train
    num_bytes: 1086515.0
    num_examples: 1196
  download_size: 338585
  dataset_size: 1086515.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---