metadata
library_name: ultralytics
tags:
- object-detection
- yolo
- yolo9
- animals
- CAN-Benchmark
license: mit
datasets:
- ICICLE-AI/CAN_Benchmark
task: object-detection
YOLOv9 – Animal Detection (Zebra, Impala, Giraffe)
This model is a YOLOv9 detector fine-tuned with Ultralytics.
It was trained for 50 epochs on a subset of the ICICLE-AI/CAN_Benchmark dataset containing three species:
- 0 → zebra
- 1 → impala
- 2 → giraffe
📊 Training details
- Framework: Ultralytics YOLOv9
- Epochs: 50
- See full hyperparameters in
args.yaml
The model converged by ~40 epochs, and shows strong precision/recall on the held-out validation set:
🚀 Usage
Load the model directly from the Hugging Face Hub:
from ultralytics import YOLO
# Load model from HF Hub
model = YOLO("ICICLE-AI/yolov9-animals-AE-data")
# Run inference
results = model("demo.jpg")
results[0].show()