Datasets:
metadata
dataset_info:
features:
- name: image
dtype: image
- name: overall_rating
dtype: float64
- name: shape_rating
dtype: float64
- name: color_rating
dtype: float64
- name: label_rating
dtype: float64
splits:
- name: train
num_bytes: 5779471.79338843
num_examples: 96
- name: test
num_bytes: 1505070.2066115702
num_examples: 25
download_size: 6301679
dataset_size: 7284542
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
task_categories:
- image-classification
size_categories:
- n<1K
Dataset Card for Dataset Name
This dataset holds 121 images of glass bottles for drinking water. The aesthetics were rated by five participants from Germany across different demographics.