Object Detection
ultralytics
Russian
yolov8
drone-detection
anti-uav
small-object-detection
Eval Results (legacy)
Instructions to use ReNothingg/drone-studio with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ultralytics
How to use ReNothingg/drone-studio with ultralytics:
from huggingface_hub import hf_hub_download from ultralytics import YOLO # pick the weights file from this repo's "Files and versions" tab weights = hf_hub_download("ReNothingg/drone-studio", "<weights>.pt") model = YOLO(weights) source = 'http://images.cocodataset.org/val2017/000000039769.jpg' model.predict(source=source, save=True) - Notebooks
- Google Colab
- Kaggle
Download studio/best.json from ReNothingg/drone-studio: direct link, hf CLI and curl.
- Browser
- Download file 1.05 kB
-
https://huggingface.co/ReNothingg/drone-studio/resolve/main/studio/best.json
- Command line
-
hf download hf://ReNothingg/drone-studio/studio/best.json
-
curl -L -o best.json https://huggingface.co/ReNothingg/drone-studio/resolve/main/studio/best.json
1.05 kB
| { | |
| "inference_confidence": 0.2982982982982983, | |
| "base": "models/hires/best.pt", | |
| "train": "runs/detect/studio_synth_v1", | |
| "data": "data/studio_synth_v1/dataset.yaml", | |
| "synthetic_val": { | |
| "mAP50": 0.995, | |
| "recall": 0.984 | |
| }, | |
| "real_val": { | |
| "mAP50": 0.894, | |
| "recall": 0.852 | |
| }, | |
| "hires_on_synthetic_val": { | |
| "mAP50": 0.572, | |
| "recall": 0.602 | |
| }, | |
| "hires_on_real_val": { | |
| "mAP50": 0.903, | |
| "recall": 0.848 | |
| }, | |
| "confidence_calibration": { | |
| "created_at": "2026-10-07T14:15:30.303141+00:00", | |
| "split": "val", | |
| "image_size": 640, | |
| "criterion": "maximum_f1", | |
| "threshold": 0.2982982982982983, | |
| "precision": 0.9330640621158387, | |
| "recall": 0.8587763545174213, | |
| "f1": 0.8943802655156419 | |
| }, | |
| "name": "Ястраб-Studio", | |
| "created_at": "2026-10-07", | |
| "architecture": "yolov8s", | |
| "image_size": 640, | |
| "source": "runs/h100/yastrab_studio/best.pt (Ястраб + 4 epochs with data/studio_synth_v1 on H100)", | |
| "sha256": "e216ce734cf7d83f18a535c1bed9fe82e7120e7bfc0477109049391d6408a4d5" | |
| } | |