Image Segmentation
ultralytics
instance-segmentation
yolo
cell-biology
microscopy
wound-healing-assay
Instructions to use nmariotto/scratch-assay-segmentation with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ultralytics
How to use nmariotto/scratch-assay-segmentation with ultralytics:
# Couldn't find a valid YOLO version tag. # Replace XX with the correct version. from ultralytics import YOLOvXX model = YOLOvXX.from_pretrained("nmariotto/scratch-assay-segmentation") source = 'http://images.cocodataset.org/val2017/000000039769.jpg' model.predict(source=source, save=True) - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -99,7 +99,8 @@ Two known limits:
|
|
| 99 |
subtracted from it.
|
| 100 |
|
| 101 |
The ceiling is not the architecture. On blinded repeat corrections the human observer
|
| 102 |
-
reproduced their own delineation at a median intersection over union of 0.861
|
|
|
|
| 103 |
the models achieve. What limits agreement here is the reproducibility of the wound boundary
|
| 104 |
itself.
|
| 105 |
|
|
|
|
| 99 |
subtracted from it.
|
| 100 |
|
| 101 |
The ceiling is not the architecture. On blinded repeat corrections the human observer
|
| 102 |
+
reproduced their own delineation at a median intersection over union of 0.861 on the 10
|
| 103 |
+
pairs that measure boundary reproducibility, and 0.894 over all 14, about what
|
| 104 |
the models achieve. What limits agreement here is the reproducibility of the wound boundary
|
| 105 |
itself.
|
| 106 |
|