metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype:
class_label:
names:
'0': duckling
'1': rubber duck
'2': unlabeled
splits:
- name: train
num_bytes: 10929870.244
num_examples: 1169
download_size: 10859997
dataset_size: 10929870.244
configs:
- config_name: default
data_files:
- split: train
path: data/train-*