Generative AI
Fine Tune Gemma
Fine-tuning Vision-Language Models (VLMs) like Google’s Gemma is one of the most exciting frontiers in AI. The ability to teach a model to see and describe the world according to a specific task is incredibly powerful. But as anyone who has tried knows, the journey from a raw dataset to a trained model is fraught with peril: massive datasets, out-of-memory errors, and tangled notebook code.
This project aims to simplify the fine-tuning process by providing a streamlined, end-to-end pipeline that takes you from data preparation to fine tuning.
Medium Article 1 Medium Medium Article 2 Medium Medium Article 3 MediumMore on Github Github
Hybrid RAG | WEBSITE CRAWLER
A Document Q&A Application with LanceDB
Streamlit-based application that enables you to ask questions about documents and receive answers based on the content of those documents. It uses Hybrid Retrieval-Augmented Generation (RAG) approach to find relevant information within the document and generate accurate, context-aware responses.
More on Github GithubText to Speech
The Text to Speech (TTS) project is a Python-based application that converts text data from an Excel file into speech audio files. The system uses a pre-trained TTS model, providing users with the ability to customize various parameters, such as the selected TTS model, encoder configuration, and speaker information.
The system integrates with the Coqui TTS library, allowing users to leverage various TTS models for speech synthesis.
More on Github GithubProject 2
Coming Soon.
Lorem ipsum dolor sit, amet consectetur adipisicing elit. Dolore, debitis commodi illo blanditiis nemo voluptatum! Quasi eligendi laborum aperiam exercitationem praesentium maxime doloribus, blanditiis tempore voluptatum delectus earum voluptates expedita!
More on Github GithubProject 3
Coming Soon.
Lorem ipsum dolor sit, amet consectetur adipisicing elit. Dolore, debitis commodi illo blanditiis nemo voluptatum! Quasi eligendi laborum aperiam exercitationem praesentium maxime doloribus, blanditiis tempore voluptatum delectus earum voluptates expedita!
More on Github GithubProject 4
Coming Soon.
Lorem ipsum dolor sit amet consectetur, adipisicing elit. Quis totam reprehenderit velit ut dolores tenetur illum, corporis, repellendus autem saepe consequuntur harum molestiae quia doloremque sit optio architecto aspernatur sint.
More on Github Github