metadata
dataset_info:
features:
- name: uid
dtype: string
- name: error_type
dtype: string
- name: good_form
dtype: string
- name: bad_form
dtype: string
- name: good_phrase
dtype: string
- name: bad_phrase
dtype: string
- name: start_index
dtype: int64
- name: upos
dtype: string
- name: lemma
dtype: string
- name: feature
dtype: string
- name: good_value
dtype: string
- name: bad_value
dtype: string
- name: good_sentence
dtype: string
- name: bad_sentence
dtype: string
- name: sentence_start_index
dtype: int64
splits:
- name: train
num_bytes: 48776153
num_examples: 79523
download_size: 7772826
dataset_size: 48776153
configs:
- config_name: default
data_files:
- split: train
path: data/train-*