metadata
dataset_info:
features:
- name: detected_obj_id
dtype: int64
- name: cropped_frame_encoded_base64
dtype: string
splits:
- name: train
num_bytes: 324924360
num_examples: 1111
download_size: 221658160
dataset_size: 324924360
configs:
- config_name: default
data_files:
- split: train
path: data/train-*