Object Detection
ultralytics
ONNX
yolo
yolo26
pcb
defect-detection
manufacturing
aoi
Eval Results (legacy)
Instructions to use steven0226/pcb-defect-detection with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ultralytics
How to use steven0226/pcb-defect-detection with ultralytics:
# Couldn't find a valid YOLO version tag. # Replace XX with the correct version. from ultralytics import YOLOvXX model = YOLOvXX.from_pretrained("steven0226/pcb-defect-detection") source = 'http://images.cocodataset.org/val2017/000000039769.jpg' model.predict(source=source, save=True) - Notebooks
- Google Colab
- Kaggle
Update ownership and portfolio links
Browse files
README.md
CHANGED
|
@@ -35,8 +35,8 @@ Ultralytics **YOLO26** (NMS-free, end-to-end detection head) fine-tuned to detec
|
|
| 35 |
bare printed-circuit-board defects: `missing_hole`, `mouse_bite`, `open_circuit`, `short`, `spur`,
|
| 36 |
`spurious_copper`.
|
| 37 |
|
| 38 |
-
- **Code, training notebooks, benchmark/ablation scripts**: [
|
| 39 |
-
- **Interactive demo**: [Space](https://huggingface.co/spaces/
|
| 40 |
|
| 41 |
## Why this matters for AOI (Automated Optical Inspection)
|
| 42 |
|
|
@@ -78,7 +78,7 @@ cite for this model's real-world generalization.
|
|
| 78 |
from huggingface_hub import hf_hub_download
|
| 79 |
from ultralytics import YOLO
|
| 80 |
|
| 81 |
-
path = hf_hub_download(repo_id="
|
| 82 |
model = YOLO(path)
|
| 83 |
results = model.predict("your_pcb_image.jpg", conf=0.25)
|
| 84 |
```
|
|
|
|
| 35 |
bare printed-circuit-board defects: `missing_hole`, `mouse_bite`, `open_circuit`, `short`, `spur`,
|
| 36 |
`spurious_copper`.
|
| 37 |
|
| 38 |
+
- **Code, training notebooks, benchmark/ablation scripts**: [source repository publication pending)
|
| 39 |
+
- **Interactive demo**: [Space](https://huggingface.co/spaces/steven0226/pcb-defect-detection)
|
| 40 |
|
| 41 |
## Why this matters for AOI (Automated Optical Inspection)
|
| 42 |
|
|
|
|
| 78 |
from huggingface_hub import hf_hub_download
|
| 79 |
from ultralytics import YOLO
|
| 80 |
|
| 81 |
+
path = hf_hub_download(repo_id="steven0226/pcb-defect-detection", filename="best.pt")
|
| 82 |
model = YOLO(path)
|
| 83 |
results = model.predict("your_pcb_image.jpg", conf=0.25)
|
| 84 |
```
|