Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,43 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: cc-by-nc-2.0
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-nc-2.0
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
# 🍝 RistoNet
|
| 6 |
+
|
| 7 |
+
**RistoNet** is your AI foodie companion! Based on **EfficientNet**, this model recognizes gourmet dishes from photos with **~91% accuracy**. Perfect for recipe apps, food blogs, or just flexing your culinary AI skills. 😎🍕
|
| 8 |
+
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
## 🚀 Features
|
| 12 |
+
- Recognizes a wide variety of gourmet dishes
|
| 13 |
+
- Lightweight and fast thanks to EfficientNet
|
| 14 |
+
- Easy Hugging Face integration
|
| 15 |
+
- Ready for research, apps, or just fun experiments
|
| 16 |
+
|
| 17 |
+
---
|
| 18 |
+
|
| 19 |
+
## 🖼️ Quick Usage
|
| 20 |
+
|
| 21 |
+
```python
|
| 22 |
+
from transformers import AutoFeatureExtractor, AutoModelForImageClassification
|
| 23 |
+
from PIL import Image
|
| 24 |
+
import torch
|
| 25 |
+
|
| 26 |
+
# Load model & feature extractor
|
| 27 |
+
extractor = AutoFeatureExtractor.from_pretrained("your-username/RistoNet")
|
| 28 |
+
model = AutoModelForImageClassification.from_pretrained("your-username/RistoNet")
|
| 29 |
+
|
| 30 |
+
# Load an image
|
| 31 |
+
image = Image.open("my_dish.jpg")
|
| 32 |
+
|
| 33 |
+
# Preprocess
|
| 34 |
+
inputs = extractor(images=image, return_tensors="pt")
|
| 35 |
+
|
| 36 |
+
# Predict
|
| 37 |
+
with torch.no_grad():
|
| 38 |
+
outputs = model(**inputs)
|
| 39 |
+
predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
|
| 40 |
+
|
| 41 |
+
# Top class
|
| 42 |
+
predicted_class = predictions.argmax().item()
|
| 43 |
+
print(f"Predicted class: {predicted_class}")
|