Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: pytorch
|
| 4 |
+
tags:
|
| 5 |
+
- computer-vision
|
| 6 |
+
- image-classification
|
| 7 |
+
- efficientnet
|
| 8 |
+
- surgical-tools
|
| 9 |
+
- surgical-instruments
|
| 10 |
+
- binary-classification
|
| 11 |
+
datasets:
|
| 12 |
+
- joonhaim/hinged-state-classifier-crops
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
# hinged-state-classifier-efficientnet-b0
|
| 16 |
+
|
| 17 |
+
EfficientNet-B0 image classification model for surgical instrument state recognition.
|
| 18 |
+
|
| 19 |
+
## Task
|
| 20 |
+
Binary classification of hinged surgical instrument state:
|
| 21 |
+
- closed
|
| 22 |
+
- open
|
| 23 |
+
|
| 24 |
+
## Model details
|
| 25 |
+
- Architecture: EfficientNet-B0
|
| 26 |
+
- Framework: PyTorch
|
| 27 |
+
- Checkpoint: `best.pt`
|
| 28 |
+
- Input image size: 224 × 224
|
| 29 |
+
|
| 30 |
+
## Dataset
|
| 31 |
+
Trained on:
|
| 32 |
+
- `joonhaim/hinged-state-classifier-crops`
|
| 33 |
+
|
| 34 |
+
## Notes
|
| 35 |
+
This is a custom PyTorch checkpoint for classifying the state of hinged surgical instruments from cropped images.
|
| 36 |
+
|
| 37 |
+
The class order must match the order used during training.
|