metadata
dataset_info:
features:
- name: image
dtype: image
- name: objects
struct:
- name: bbox
sequence:
sequence: float64
- name: category
sequence: int64
splits:
- name: train
num_bytes: 13646239
num_examples: 322
- name: test
num_bytes: 3924231
num_examples: 93
download_size: 17424795
dataset_size: 17570470
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
license: cc-by-4.0
task_categories:
- object-detection
language:
- en
tags:
- medical
size_categories:
- n<1K