metadata
dataset_info:
features:
- name: SeqA
dtype: string
- name: SeqB
dtype: string
- name: labels
dtype: int64
splits:
- name: train
num_bytes: 81454
num_examples: 288
download_size: 7355
dataset_size: 81454
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
Parsed data from the PPI challenge 1 out of the Dickinson Group.
All credit goes to them for putting together such a great initiative. You can find the challenge webpage here.