---
license: apache-2.0
dataset_info:
  features:
  - name: noisy
    dtype: large_string
  - name: gt
    dtype: large_string
  - name: type
    dtype: large_string
  splits:
  - name: train
    num_bytes: 3336936763
    num_examples: 15000000
  - name: test
    num_bytes: 111594178
    num_examples: 500000
  download_size: 801424830
  dataset_size: 3448530941
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---