metadata
dataset_info:
features:
- name: IdA
dtype: string
- name: IdB
dtype: string
- name: labels
dtype: int64
- name: mechanism
dtype: string
- name: effect
dtype: string
- name: score
dtype: float64
- name: sentence
dtype: string
- name: signor_id
dtype: string
splits:
- name: train
num_bytes: 8075990
num_examples: 20260
download_size: 3280562
dataset_size: 8075990
configs:
- config_name: default
data_files:
- split: train
path: data/train-*