metadata
dataset_info:
features:
- name: condition_type
sequence: string
- name: condition_value
sequence: string
- name: target_type
dtype: string
- name: target_candidates
list:
- name: count
dtype: float64
- name: value
dtype: string
- name: keywords_type
dtype: string
- name: hash
dtype: string
- name: question
dtype: string
- name: options
sequence: string
- name: option_weights
sequence: float64
splits:
- name: train
num_bytes: 712764
num_examples: 556
download_size: 295036
dataset_size: 712764
configs:
- config_name: default
data_files:
- split: train
path: data/train-*