metadata
dataset_info:
features:
- name: word_lines
dtype: string
- name: label_lines
sequence: int64
- name: merge_error
dtype: bool
splits:
- name: train
num_bytes: 1362372233
num_examples: 1801350
- name: validation
num_bytes: 2879764
num_examples: 3760
- name: test
num_bytes: 3245389
num_examples: 4358
download_size: 395896629
dataset_size: 1368497386
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*