README.md

metadata

title: FoodVision Big
emoji: 🚀
colorFrom: pink
colorTo: blue
sdk: gradio
sdk_version: 5.32.1
app_file: app.py
pinned: false
license: mit

🍔 FoodVision Big 🚀

Welcome to FoodVision Big, a computer vision web app powered by EfficientNetB2 trained on the Food101 dataset.

Upload a photo of any dish and get instant predictions of what it might be — from sushi to spaghetti, from cheesecake to chow mein.

---

🧠 Model Info

• Architecture: EfficientNetB2
• Training Style: Feature extraction
• Dataset: Food101
• Classes: 101 different food types 🍱🍕🍜
• Trained On: Google Colab with PyTorch
• Accuracy: ~65% top-1

---

🌍 Try the App

🚀 Click here to launch the app

---

💻 How to Use

1. Upload a photo of a food dish.
2. Wait a few seconds (depending on GPU availability).
3. View the top 5 predicted food types with confidence scores.

---

🔧 Tech Stack

• Gradio for the web interface
• PyTorch for model training and inference
• Hugging Face Spaces for deployment
• EfficientNetB2 for accurate, efficient image classification

---

💡 Inspiration

Built as part of my deep learning journey through the "Zero to Mastery PyTorch" course by Daniel Bourke.

This project represents my first large model deployment. I'm proud of how far it's come — and there's more to come. 💪

---

🛠️ Future Improvements

• Upgrade model to fine-tuned version or use transfer learning
• Add GPU support for faster predictions
• Add support for multilingual UI
• Deploy mobile-friendly version
• Add more detailed nutritional info (just for fun 😉)

---

📜 License

This project is licensed under the MIT License.

---

🖤 From Me to You

I built this with limited resources, a lot of curiosity, and a dream to do something real. If you're learning too — keep going. Even messy progress is progress.

“Little me would be proud. Big me is getting there.” 🌟

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference