metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype:
class_label:
names:
'0': '0'
'1': '1'
'2': '2'
'3': '3'
'4': '4'
splits:
- name: train
num_bytes: 4792256325.5
num_examples: 35125
download_size: 4807906176
dataset_size: 4792256325.5
configs:
- config_name: default
data_files:
- split: train
path: data/train-*