Create README.md

README.md

@@ -1,12 +1,135 @@
 ---
-title: IngredientClassification
-emoji: 👀
-colorFrom: yellow
-colorTo: indigo
-sdk: gradio
-sdk_version: 6.2.0
-app_file: app.py
-pinned: false
+license: mit
+tags:
+- food
+- ingredient-classification
+- computer-vision
+- multi-label-classification
+- pytorch
+library_name: pytorch
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Food Ingredient Classifier
+
+A multi-label classification model that identifies ingredients in food images.
+
+## Model Description
+
+This model uses a ResNet-50 backbone fine-tuned on the Food-101 dataset to classify 101 different food ingredients.
+
+## Performance
+
+- **Validation Accuracy**: 99.01%
+- **Architecture**: ResNet-50 with custom classification head
+- **Training Epochs**: 1
+- **Number of Classes**: 101
+
+## Ingredients Classified
+
+apple_pie, baby_back_ribs, baklava, beef_carpaccio, beef_tartare, beet_salad, beignets, bibimbap, bread_pudding, breakfast_burrito, bruschetta, caesar_salad, cannoli, caprese_salad, carrot_cake, ceviche, cheese_plate, cheesecake, chicken_curry, chicken_quesadilla, chicken_wings, chocolate_cake, chocolate_mousse, churros, clam_chowder, club_sandwich, crab_cakes, creme_brulee, croque_madame, cup_cakes, deviled_eggs, donuts, dumplings, edamame, eggs_benedict, escargots, falafel, filet_mignon, fish_and_chips, foie_gras, french_fries, french_onion_soup, french_toast, fried_calamari, fried_rice, frozen_yogurt, garlic_bread, gnocchi, greek_salad, grilled_cheese_sandwich, grilled_salmon, guacamole, gyoza, hamburger, hot_and_sour_soup, hot_dog, huevos_rancheros, hummus, ice_cream, lasagna, lobster_bisque, lobster_roll_sandwich, macaroni_and_cheese, macarons, miso_soup, mussels, nachos, omelette, onion_rings, oysters, pad_thai, paella, pancakes, panna_cotta, peking_duck, pho, pizza, pork_chop, poutine, prime_rib, pulled_pork_sandwich, ramen, ravioli, red_velvet_cake, risotto, samosa, sashimi, scallops, seaweed_salad, shrimp_and_grits, spaghetti_bolognese, spaghetti_carbonara, spring_rolls, steak, strawberry_shortcake, sushi, tacos, takoyaki, tiramisu, tuna_tartare, waffles
+
+## Usage
+
+```python
+import torch
+from torchvision import transforms, models
+from PIL import Image
+import torch.nn as nn
+from huggingface_hub import hf_hub_download
+
+# Model architecture
+class FoodIngredientClassifier(nn.Module):
+    def __init__(self, num_classes=101):
+        super().__init__()
+        self.backbone = models.resnet50(pretrained=False)
+        num_features = self.backbone.fc.in_features
+        self.backbone.fc = nn.Sequential(
+            nn.Dropout(0.5),
+            nn.Linear(num_features, 512),
+            nn.ReLU(),
+            nn.Dropout(0.3),
+            nn.Linear(512, num_classes)
+        )
+    
+    def forward(self, x):
+        return self.backbone(x)
+
+# Load model
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+model = FoodIngredientClassifier(num_classes=101).to(device)
+
+# Download checkpoint from Hugging Face
+checkpoint_path = hf_hub_download(
+    repo_id="YOUR_USERNAME/food-ingredient-classifier",
+    filename="best_model.pth"
+)
+checkpoint = torch.load(checkpoint_path, map_location=device, weights_only=False)
+model.load_state_dict(checkpoint['model_state_dict'])
+model.eval()
+
+# Get ingredient labels
+mlb = checkpoint['mlb']
+
+# Prepare image
+transform = transforms.Compose([
+    transforms.Resize((224, 224)),
+    transforms.ToTensor(),
+    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
+])
+
+# Inference
+img = Image.open('food_image.jpg').convert('RGB')
+img_tensor = transform(img).unsqueeze(0).to(device)
+
+with torch.no_grad():
+    output = model(img_tensor)
+    probs = torch.sigmoid(output).cpu().numpy()[0]
+
+# Get predictions (threshold at 0.5)
+threshold = 0.5
+pred_indices = (probs > threshold).nonzero()[0]
+ingredients = mlb.classes_[pred_indices]
+confidences = probs[pred_indices]
+
+for ingredient, confidence in zip(ingredients, confidences):
+    print(f"{ingredient}: {confidence:.2%}")
+```
+
+## Training Details
+
+- **Dataset**: Food-101 (101,000 images across 101 categories)
+- **Batch Size**: 64
+- **Optimizer**: AdamW with weight decay (0.01)
+- **Learning Rate**: 0.001 with OneCycleLR scheduler
+- **Data Augmentation**: Random horizontal flip, rotation (15°), color jitter
+- **Mixed Precision Training**: Enabled for faster training
+- **Image Size**: 224x224
+
+## Model Architecture
+
+- **Backbone**: ResNet-50 (pretrained on ImageNet)
+- **Custom Head**:
+  - Dropout (0.5)
+  - Linear (2048 → 512)
+  - ReLU
+  - Dropout (0.3)
+  - Linear (512 → 101)
+- **Output**: Multi-label (sigmoid activation)
+
+## Citation
+
+If you use this model, please cite:
+
+```bibtex
+@misc{food-ingredient-classifier,
+  author = {Your Name},
+  title = {Food Ingredient Classifier},
+  year = {2025},
+  publisher = {Hugging Face},
+  howpublished = {\url{https://huggingface.co/YOUR_USERNAME/food-ingredient-classifier}}
+}
+```
+
+## License
+
+MIT License