metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype:
class_label:
names:
'0': double_bottom
'1': double_top
'2': head_and_shoulders
'3': random
splits:
- name: train
num_bytes: 1179842
num_examples: 400
download_size: 1026180
dataset_size: 1179842
configs:
- config_name: default
data_files:
- split: train
path: data/train-*