---
dataset_info:
  features:
  - name: prediction
    dtype: string
  - name: 'true'
    dtype: string
  - name: index
    dtype: int64
  splits:
  - name: pred
    num_bytes: 100113
    num_examples: 1000
  download_size: 29404
  dataset_size: 100113
configs:
- config_name: default
  data_files:
  - split: pred
    path: data/pred-*
---