metadata
dataset_info:
features:
- name: document_id
dtype: string
- name: document_text
dtype: string
- name: title
dtype: string
- name: abstract
dtype: string
- name: annotation_ids
list: string
- name: annotation_texts
list: string
- name: annotation_offsets
list: uint32
- name: annotation_lengths
list: uint32
- name: annotation_groundings
list: string
- name: annotation_types
list: string
- name: relation_entity1
list: string
- name: relation_entity2
list: string
- name: relation_type
list: string
- name: relation_novel
list: string
splits:
- name: train
num_bytes: 2344285
num_examples: 400
- name: validation
num_bytes: 621048
num_examples: 100
- name: test
num_bytes: 608756
num_examples: 100
download_size: 1503150
dataset_size: 3574089
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*