metadata
language:
- vi
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
- name: embeddings
sequence: float32
splits:
- name: train
num_bytes: 120681954
num_examples: 898
download_size: 121588411
dataset_size: 120681954
configs:
- config_name: default
data_files:
- split: train
path: data/train-*