metadata
dataset_info:
features:
- name: stop_phrase
dtype: string
- name: doc_freq
dtype: int64
- name: ttf
dtype: int64
- name: term_freq
dtype: int64
- name: tokens
list:
- name: end_offset
dtype: int64
- name: position
dtype: int64
- name: start_offset
dtype: int64
- name: score
dtype: float64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 5101485
num_examples: 29740
download_size: 2756995
dataset_size: 5101485
configs:
- config_name: default
data_files:
- split: train
path: data/train-*