Downloads · 30 days
0
basketball-rf-detr-m-640
ortizeg/basketball-rf-detr-m-640
basketball-rf-detr-m-640 is a object detection model from ortizeg. Use it when you need objects located in an image. It is set up for onnx. The card lists the license as apache-2.0.
RF-DETR-Medium, a real-time transformer detector fine-tuned on the basketball dataset at 640x640, using the rfdetr-standard top-k=300 multi-label decode required to match its native COCOeval score.
Access
Public
Updated Jul 30, 2026
Repo size
128 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.onnx128 MB · 100%
At a glance
- Task
- Object Detection
- Library
- onnx
- License
- apache-2.0
- Access
- Public
- Created
- Jul 30, 2026
- Updated
- Jul 30, 2026
- SHA
- ffc7d08a
Model card
From the Hugging Face model README
rf-detr-m-640
RF-DETR-Medium, a real-time transformer detector fine-tuned on the basketball dataset at 640x640, using the rfdetr-standard top-k=300 multi-label decode required to match its native COCOeval score.
Metrics
Measured on basketball-player-detection-3 (test split), via the object-detection-eval harness.
| Metric | 5-class | 10-class |
|---|---|---|
| mAP@50:95 | 0.646 | 0.590 |
| mAP@50 | 0.937 | 0.812 |
Preprocessing
- Resize:
square(alignment:none) - Normalize:
mean_std - Channel order:
RGB - Input shape:
(3, 640, 640)(float32)
Provenance
- Source repo: https://github.com/roboflow/rf-detr
- Training config:
docs/provenance/configs/rfdetr_m/train_rfdetr.py - Hardware: vast.ai A100 80GB (single GPU)
- Command:
python train_rfdetr.py (rfdetr==1.5.2, RFDETRMedium.train(); see docs/provenance/configs/rfdetr_m/train_rfdetr.py)
Usage
This ONNX file is one of the 7-model roster benchmarked in object-detection-eval. Load it through the registry for verified, hash-checked download and the exact preprocessing recipe above:
from object_detection_eval.registry import ModelRegistry, download_weights
registry = ModelRegistry.from_directory("registry")
card = registry.get("rf-detr-m-640")
weights_path = download_weights(card)
- Task
- Object Detection
- Library
- onnx
- License
- apache-2.0
- Created
- Jul 30, 2026
- Updated
- Jul 30, 2026