Downloads · 30 days
0
basketball-damo-yolo-m-640
ortizeg/basketball-damo-yolo-m-640
basketball-damo-yolo-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.
DAMO-YOLO-M (tinynasL35M backbone) fine-tuned on the basketball dataset at 640x640. Apache-2.0 CNN detector; underperformed its COCO pedigree (49.2 mAP@50:95) on this 465-image basketball set relative to the transform…
Access
Public
Updated Jul 30, 2026
Repo size
111 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.onnx111 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
- 1ca20f40
Model card
From the Hugging Face model README
damo-yolo-m-640
DAMO-YOLO-M (tinynasL35_M backbone) fine-tuned on the basketball dataset at 640x640. Apache-2.0 CNN detector; underperformed its COCO pedigree (49.2 mAP@50:95) on this 465-image basketball set relative to the transformer models in the roster.
Metrics
Measured on basketball-player-detection-3 (test split), via the object-detection-eval harness.
| Metric | 5-class | 10-class |
|---|---|---|
| mAP@50:95 | 0.619 | 0.541 |
| mAP@50 | 0.890 | 0.787 |
Preprocessing
- Resize:
square(alignment:none) - Normalize:
none - Channel order:
RGB - Input shape:
(3, 640, 640)(float32)
Provenance
- Source repo: https://github.com/tinyvision/DAMO-YOLO
- Training config:
docs/provenance/configs/damo_m/damoyolo_basketball_m.py - Hardware: vast.ai RTX 3090 (single GPU)
- Command:
PYTHONPATH=. python -m torch.distributed.launch --nproc_per_node=1 tools/train.py -f docs/provenance/configs/damo_m/damoyolo_basketball_m.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("damo-yolo-m-640")
weights_path = download_weights(card)
- Task
- Object Detection
- Library
- onnx
- License
- apache-2.0
- Created
- Jul 30, 2026
- Updated
- Jul 30, 2026