metadata
dataset_info:
features:
- name: phrase1
dtype: string
- name: phrase2
dtype: string
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: label
dtype:
class_label:
names:
'0': negative
'1': positive
- name: idx
dtype: int32
splits:
- name: train
num_bytes: 2064490
num_examples: 7004
- name: validation
num_bytes: 293434
num_examples: 1000
- name: test
num_bytes: 588478
num_examples: 2000
download_size: 1997180
dataset_size: 2946402
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*