Instructions to use roboseasylabs/red_cube_yolo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ultralytics
How to use roboseasylabs/red_cube_yolo with ultralytics:
# Couldn't find a valid YOLO version tag. # Replace XX with the correct version. from ultralytics import YOLOvXX model = YOLOvXX.from_pretrained("roboseasylabs/red_cube_yolo") source = 'http://images.cocodataset.org/val2017/000000039769.jpg' model.predict(source=source, save=True) - Notebooks
- Google Colab
- Kaggle
red_cube_yolo
Ultralytics YOLO κ°μ²΄ νμ§ λͺ¨λΈμ
λλ€. νμ΅ μ€ν cube μ κ²°κ³Όμ
λλ€.
- ν΄λμ€ (1κ°):
cube - μ λ ₯ ν΄μλ: 640 x 640
- λ² μ΄μ€ λͺ¨λΈ:
yolo26n.pt
μ±λ₯
79 epoch νμ΅ ν validation κ²°κ³Όμ λλ€.
| μ§ν | κ° |
|---|---|
| Precision | 0.9100 |
| Recall | 0.9275 |
| mAP@50 | 0.9657 |
| mAP@50-95 | 0.9637 |
μ¬μ©λ²
pip install ultralytics huggingface_hub
from huggingface_hub import hf_hub_download
from ultralytics import YOLO
weights = hf_hub_download("roboseasylabs/red_cube_yolo", "best.pt")
model = YOLO(weights)
results = model.predict("image.jpg", conf=0.25, iou=0.45)
results[0].show()
νμ΅ μ€μ
| νλͺ© | κ° |
|---|---|
| λ² μ΄μ€ κ°μ€μΉ | yolo26n.pt |
| epochs | 200 |
| imgsz | 640 |
| batch | 16 |
| patience | 20 |
λ°μ΄ν°μ Β· νμ΄νλΌμΈ
- λ°μ΄ν°μ : https://universe.roboflow.com/s-workspace-7c6k7/red_cube-rjoad
- νμ΅ νμ΄νλΌμΈ(촬μ β νλ μ λΆν β λΌλ²¨λ§ β νμ΅ β μΆλ‘ ): https://github.com/roboseasy/YOLO
νκ³
- λ¨μΌ νκ²½μμ μμ§ν λ°μ΄ν°λ‘ νμ΅νμ΅λλ€. λ°°κ²½Β·μ‘°λͺ Β·μΉ΄λ©λΌκ° λ°λλ©΄ μ±λ₯μ΄ λ¨μ΄μ§ μ μμ΅λλ€.
- 640x640 μ λ ₯ κΈ°μ€μ λλ€. μμ£Ό μκ² μ°ν 물체λ λμΉ μ μμ΅λλ€.
- Downloads last month
- 68
Evaluation results
- Precisionself-reported0.910
- Recallself-reported0.927
- mAP@50self-reported0.966
- mAP@50-95self-reported0.964