metadata
dataset_info:
features:
- name: pmcid
dtype: string
- name: title
dtype: string
- name: abstract
dtype: string
- name: fulltext
dtype: string
- name: file_path
dtype: string
splits:
- name: train
num_bytes: 78534181
num_examples: 2989
download_size: 40500470
dataset_size: 78534181
configs:
- config_name: default
data_files:
- split: train
path: data/train-*