metadata
dataset_info:
features:
- name: generation_output_media_id
dtype: string
- name: generation_output_media
dtype: image
- name: review_stage
dtype: int8
- name: review_status
dtype: string
- name: accuracy_reasons
sequence: string
- name: accuracy_notes
dtype: string
- name: aesthetics_reasons
sequence: string
- name: aesthetics_notes
dtype: string
- name: distinctiveness_reasons
sequence: string
- name: distinctiveness_notes
dtype: string
- name: restaurant_item_id
dtype: string
- name: restaurant_item_name
dtype: string
- name: restaurant_item_description
dtype: string
- name: ref_descriptions
list:
- name: id
dtype: string
- name: restaurant_item_id
dtype: string
- name: source_name
dtype: string
- name: source_description
dtype: string
- name: ref_images
list:
- name: id
dtype: string
- name: restaurant_item_id
dtype: string
- name: media_id
dtype: string
- name: media
dtype: image
splits:
- name: training
num_bytes: 1846909990
num_examples: 598
- name: testing
num_bytes: 526382350
num_examples: 198
download_size: 2320109699
dataset_size: 2373292340
configs:
- config_name: default
data_files:
- split: training
path: data/training-*
- split: testing
path: data/testing-*