metadata
dataset_info:
features:
- name: sentence
sequence: string
- name: summary
sequence: string
- name: oracle
sequence: int64
- name: label_list
sequence: int64
- name: scores
struct:
- name: ROUGE
dtype: float64
splits:
- name: train
num_bytes: 948448045
num_examples: 287113
- name: validation
num_bytes: 43312404
num_examples: 13368
- name: test
num_bytes: 37130119
num_examples: 11490
download_size: 609207530
dataset_size: 1028890568
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*