metadata
dataset_info:
features:
- name: image
dtype: image
- name: labels
dtype:
class_label:
names:
'0': Poor
'1': Fair
'2': Good
'3': Great
'4': Excellent
'5': Not Applicable
splits:
- name: train
num_bytes: 17149258680.978
num_examples: 36626
- name: test
num_bytes: 985364854.734
num_examples: 1998
- name: validation
num_bytes: 1897320754.645
num_examples: 3979
download_size: 19146234501
dataset_size: 20031944290.357002
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*