Skip to content

Explore the rich flavors of Indian desserts with TunedLlavaDelights. Utilizing the in Llava fine-tuning, our project unveils detailed nutritional profiles, taste notes, and optimal consumption times for beloved sweets. Dive into a fusion of AI innovation and culinary tradition

Notifications You must be signed in to change notification settings

Ravi-Teja-konda/TunedLlavaDelights

Repository files navigation

TunedLlavaDelights : Llava Fine-Tune on Indian Desserts with Nutritional Info

Overview

This project is an exploration into the capabilities of fine-tuning the Llava model specifically for Indian desserts. We've successfully fine-tuned the model to identify and provide in-depth nutritional information on 5 distinct Indian dessert items. This not only includes a brief overview of each dessert, covering aspects like taste and nutritional content, but also allergen notes, recommended times to eat, and feel after consumption.

🎥 Demo Video

Watch our model in action and see how it beautifully integrates AI with culinary tradition:

Demo Video

🌟 Key Highlights:

  • Enhanced Detection: Our model can recognize some dessert items that were previously undetectable by standard Llava and even GPT-4 models.
  • Image Recognition: It showcases improved accuracy in image detection, surpassing GPT-4v in recognizing specific images and providing detailed information.
  • Replicate Training: Utilized Replicate for training and hosting, ensuring a seamless integration and user experience.

Model Link 📎

Experience the model firsthand and explore its capabilities: Llava Fine-Tuned Model on Replicate

Purpose 💡

This project serves as a dynamic showcase of how fine-tuning LVM (Language Vision Models) on domain-specific datasets can significantly enhance AI's understanding and analysis capabilities.

Applications 🚀

  • PCB Verification: Training LVMs on images of PCBs (Printed Circuit Boards) to verify and filter them based on specific queries.
  • Domain-Specific Training: Tailoring the model to recognize and evaluate domain-specific or private image datasets for more effective and accurate recognition.

Getting Started 🌟

To get started with this project and explore the fine-tuned Llava model's capabilities, please follow the links provided above. Your journey into the advanced application of AI in understanding and providing detailed insights into Indian desserts begins here!

References

Contribution 🤝

Contributions, questions, and feedback are welcome! Feel free to open an issue or submit a pull request.

About

Explore the rich flavors of Indian desserts with TunedLlavaDelights. Utilizing the in Llava fine-tuning, our project unveils detailed nutritional profiles, taste notes, and optimal consumption times for beloved sweets. Dive into a fusion of AI innovation and culinary tradition

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages