---
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
dataset_info:
  features:
  - name: image
    dtype:
      image:
        decode: false
  - name: transcription
    dtype: string
  - name: sample_id
    dtype: string
  - name: source
    dtype: string
  - name: metadata
    dtype: string
  splits:
  - name: train
    num_bytes: 62564294301
    num_examples: 300000
  download_size: 61811590195
  dataset_size: 62564294301
---