Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- image-classification
|
| 4 |
+
- pytorch
|
| 5 |
+
- medical
|
| 6 |
+
library_name: timm
|
| 7 |
+
classes:
|
| 8 |
+
- Chickenpox
|
| 9 |
+
- Cowpox
|
| 10 |
+
- HFMD
|
| 11 |
+
- Healthy
|
| 12 |
+
- Measles
|
| 13 |
+
- Monkeypox
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
# Mpox Detection Model
|
| 17 |
+
|
| 18 |
+
Ce modèle est un Vision Transformer (ViT) finetuné pour détecter différentes maladies de la peau, notamment la variole du singe (Mpox).
|
| 19 |
+
|
| 20 |
+
## Classes
|
| 21 |
+
Le modèle peut classifier les images dans les catégories suivantes :
|
| 22 |
+
- Chickenpox (Varicelle)
|
| 23 |
+
- Cowpox (Variole bovine)
|
| 24 |
+
- HFMD (Syndrome pieds-mains-bouche)
|
| 25 |
+
- Healthy (Sain)
|
| 26 |
+
- Measles (Rougeole)
|
| 27 |
+
- Monkeypox (Variole du singe)
|
| 28 |
+
|
| 29 |
+
## Usage
|
| 30 |
+
|
| 31 |
+
```python
|
| 32 |
+
import torch
|
| 33 |
+
from models import create_model
|
| 34 |
+
from torchvision import transforms
|
| 35 |
+
from PIL import Image
|
| 36 |
+
|
| 37 |
+
# Charger le modèle
|
| 38 |
+
model = create_model(pretrained=False)
|
| 39 |
+
model.load_state_dict(torch.load("best_mpox_model.pth"))
|
| 40 |
+
model.eval()
|
| 41 |
+
|
| 42 |
+
# Prétraitement
|
| 43 |
+
transform = transforms.Compose([
|
| 44 |
+
transforms.Resize((224, 224)),
|
| 45 |
+
transforms.ToTensor(),
|
| 46 |
+
transforms.Normalize([0.5]*3, [0.5]*3)
|
| 47 |
+
])
|
| 48 |
+
|
| 49 |
+
# Prédiction
|
| 50 |
+
img = Image.open("path/to/image.jpg")
|
| 51 |
+
img_t = transform(img).unsqueeze(0)
|
| 52 |
+
output = model(img_t)
|
| 53 |
+
predicted_idx = output.argmax(1).item()
|
| 54 |
+
print(predicted_idx)
|
| 55 |
+
```
|