the-khiem7/snakeaid-yolov12-5000-bbox

SnakeAid YOLOv12 5000 BBox

Dataset Summary

This repository contains a YOLO-format SnakeAid object-detection dataset for snake detection experiments. It is organized as image/label pairs across train, valid, test splits and is intended for training or evaluating YOLO-family detectors, including the related SnakeAid Detect YOLOv12 checkpoints linked below.

Safety note: snake detection can be safety-critical in real-world use. Treat model outputs trained on this data as assistive signals only; do not use them as the sole basis for handling, approaching, or identifying a snake.

Key Details

Field	Value
Format	YOLO object detection
Splits	train, valid, test
Images	5291
Label files	5291
Classes	22
License metadata	cc0-1.0
Uploadable local files	21172

Splits

Split	Images	Labels
train	4795	4795
valid	265	265
test	231	231

File Layout

data.yaml
train/images/*.jpg
train/labels/*.txt
valid/images/*.jpg
valid/labels/*.txt
test/images/*.jpg
test/labels/*.txt

Each image is paired with a YOLO .txt label file using the same stem. Class names and split paths are defined in data.yaml.

Classes

Class ID	Name
0	cap_nia_bac
1	cap_nia_nam
2	cap_nong
3	ho_mang_chua
4	ho_mang_xiem
5	khiem_vach
6	luc_cuom
7	luc_nua
8	luc_xanh
9	luc_xanh_duoi_do
10	ran_cuom
11	ran_dai_lon
12	ran_hoa_can_van_dom
13	ran_hoa_co_do
14	ran_rao
15	ran_rao_trau
16	ran_ri_ca
17	ran_roi
18	ran_sai_co
19	ran_soc_dua
20	ran_soc_go
21	ran_trun

Loading Example

from huggingface_hub import snapshot_download

dataset_dir = snapshot_download(
    repo_id="the-khiem7/snakeaid-yolov12-5000-bbox",
    repo_type="dataset",
)
print(dataset_dir)

For YOLO training, point your training command at the downloaded data.yaml.

Provenance

• workspace: anh-by-i-n-nh-em
• project: my-first-project-hyskh
• version: 6
• license: Public Domain
• url: https://universe.roboflow.com/anh-by-i-n-nh-em/my-first-project-hyskh/dataset/6

Related Models

• the-khiem7/snakeaid-detect-yolov12s-17e-5000bbox
• the-khiem7/snakeaid-detect-yolov12-v4-5000bbox

Limitations

• The dataset is provided as a local YOLO export, not as a fully curated benchmark.
• Class balance, duplicate images, annotation quality, and real-world geographic coverage have not been independently audited in this upload workflow.
• Performance can vary significantly with lighting, camera angle, occlusion, species similarity, and image quality.
• Use additional validation before deploying a detector trained on this data in field or safety-sensitive settings.