metadata
dataset_info:
features:
- name: article
dtype: string
- name: summary
dtype: string
- name: article_len_approx
dtype: int64
- name: summary_len_approx
dtype: int64
- name: orig_idx
dtype: int64
- name: summary_entities
list:
- name: end
dtype: int64
- name: entity
dtype: string
- name: index
dtype: int64
- name: score
dtype: float32
- name: start
dtype: int64
- name: word
dtype: string
- name: article_entities
list:
- name: end
dtype: int64
- name: entity
dtype: string
- name: index
dtype: int64
- name: score
dtype: float32
- name: start
dtype: int64
- name: word
dtype: string
splits:
- name: train
num_bytes: 2491767354
num_examples: 3783821
- name: validation
num_bytes: 125101358
num_examples: 188811
- name: test
num_bytes: 1179245
num_examples: 1822
download_size: 1011521359
dataset_size: 2618047957
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*