metadata
dataset_info:
features:
- name: text
dtype: string
- name: label
dtype:
class_label:
names:
'0': '0'
'1': '1'
'2': '2'
- name: group_idx
dtype: int64
- name: node_idx
dtype: int64
- name: train_0
dtype: bool
- name: validation_0
dtype: bool
- name: test_0
dtype: bool
splits:
- name: train
num_bytes: 259289977.0675052
num_examples: 156000
download_size: 145171506
dataset_size: 259289977.0675052
configs:
- config_name: default
data_files:
- split: train
path: data/train-*