metadata
dataset_info:
features:
- name: good_text
dtype: string
- name: bad_text
dtype: string
- name: error_family
dtype: string
- name: error_type
dtype: string
- name: uid
dtype: string
- name: misc
dtype: string
- name: error_class
dtype: int64
splits:
- name: train
num_bytes: 47851250
num_examples: 104558
- name: test
num_bytes: 11960629
num_examples: 26139
download_size: 10045986
dataset_size: 59811879
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*