metadata
dataset_info:
features:
- name: img
dtype: image
- name: image_embeddings
sequence: float64
splits:
- name: train
num_bytes: 103447384
num_examples: 1000
download_size: 103650691
dataset_size: 103447384
configs:
- config_name: default
data_files:
- split: train
path: data/train-*