metadata
dataset_info:
features:
- name: id
dtype: int64
- name: db_id
dtype: string
- name: schema
dtype: string
- name: question
dtype: string
- name: output
dtype: string
- name: evidence
dtype: string
- name: filtered_schema
sequence: string
splits:
- name: train
num_bytes: 50330088
num_examples: 9428
- name: valid
num_bytes: 6387832
num_examples: 1534
download_size: 1699011
dataset_size: 56717920
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*