metadata
dataset_info:
features:
- name: media_to_review_id
dtype: string
- name: media_to_review
dtype: image
- name: questions
sequence: string
- name: answers
sequence: string
- name: item_name
dtype: string
- name: item_description
dtype: string
- name: reference_media
sequence: image
- name: review_result
dtype: string
- name: accuracy_reasons
sequence: string
- name: aesthetics_reasons
sequence: string
- name: distinctiveness_reasons
sequence: string
- name: accuracy_notes
dtype: string
- name: aesthetics_notes
dtype: string
- name: distinctiveness_notes
dtype: string
- name: comments
dtype: string
- name: latest_review
dtype: string
splits:
- name: train
num_bytes: 4641628845.507
num_examples: 2959
- name: test
num_bytes: 1124724392
num_examples: 740
download_size: 5352813645
dataset_size: 5766353237.507
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*