metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*
dataset_info:
features:
- name: patent_num
dtype: int64
- name: claim_num1
dtype: int64
- name: claim_num2
dtype: int64
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: label
dtype: float64
splits:
- name: train
num_bytes: 1141724170.7014475
num_examples: 453043
- name: test
num_bytes: 63431500.71087167
num_examples: 25170
- name: valid
num_bytes: 63428980.58768093
num_examples: 25169
download_size: 479868966
dataset_size: 1268584652