metadata
dataset_info:
features:
- name: image_path
dtype: image
- name: text
dtype: string
- name: label_id
dtype: int32
- name: filename
dtype: string
splits:
- name: train
num_bytes: 181142299
num_examples: 85270
- name: validation
num_bytes: 25384261
num_examples: 11878
- name: test
num_bytes: 41790721
num_examples: 19635
download_size: 220793428
dataset_size: 248317281
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*