metadata
dataset_info:
features:
- name: image
dtype: image
- name: smiles
dtype: string
- name: caption
dtype: string
- name: selfies
dtype: string
splits:
- name: train
num_bytes: 11301133178.603703
num_examples: 634315
- name: validation
num_bytes: 3015567069.01932
num_examples: 168475
download_size: 2903953421
dataset_size: 14316700247.623022
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*