metadata
dataset_info:
features:
- name: pmcid
dtype: string
- name: pmid
dtype: string
- name: topic
dtype: string
- name: confidence
dtype: float32
- name: class_distribution
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 16092306545
num_examples: 10313320
- name: validation
num_bytes: 100939777
num_examples: 51377
- name: test
num_bytes: 100308273
num_examples: 51378
download_size: 8899428088
dataset_size: 16293554595
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*