metadata
dataset_info:
features:
- name: id
dtype: int64
- name: text
dtype: string
- name: attributes
sequence: string
- name: values
sequence: string
- name: values_indices
sequence:
sequence: int64
- name: values_text
dtype: string
- name: attributes_values
dtype: string
- name: json_answer
dtype: string
- name: candidate_attributes
sequence: string
- name: candidate_text
dtype: string
- name: candidate_example
struct:
- name: json_answer
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 13902395
num_examples: 31604
- name: validation
num_bytes: 1731774
num_examples: 3950
- name: test
num_bytes: 3270077
num_examples: 3951
download_size: 6561072
dataset_size: 18904246
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*