metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: unsupervised
path: data/unsupervised-*
dataset_info:
features:
- name: text
dtype: string
- name: label
dtype:
class_label:
names:
'0': neg
'1': pos
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 97632823
num_examples: 25000
- name: test
num_bytes: 96850685
num_examples: 25000
- name: unsupervised
num_bytes: 195506794
num_examples: 50000
download_size: 135785876
dataset_size: 389990302