metadata
dataset_info:
features:
- name: image
dtype: image
- name: image_id
dtype: int32
- name: abundance_type
dtype: string
- name: bboxes
sequence:
sequence: float32
splits:
- name: train
num_bytes: 78938224
num_examples: 30
download_size: 78266712
dataset_size: 78938224
configs:
- config_name: default
data_files:
- split: train
path: data/train-*