Llama2-Medical-Chatbot is a medical chatbot that uses the Llama-2-7B-Chat-GGML model and the pdf The Gale Encyclopedia of Medicine, Volume 1, 2nd Edition. It is still under development, but it has the potential to be a valuable tool for patients, healthcare professionals, and researchers.

The Llama-2-GGML-CSV-Chatbot is a conversational tool leveraging the powerful Llama-2 7B language model. It facilitates multi-turn interactions based on uploaded CSV data, allowing users to engage in seamless conversations.

To make LLM faster we need faster retrieval system. Here comes Embedding Quantization. Embedding quantization is great technique to save cost on Vector DB, significantly faster retrieval while preserving retrieval performance.

MediChat: An AI-powered medical chatbot using the Llama-2-7B-Chat model for precise clinical responses. Integrates Chroma DB and all-MiniLM-L6-v2 embeddings trained on medical literature, including texts like Clinical Emergency Medicine and Gale Encyclopedia. Accurate, fast, and reliable for healthcare queries.

NLP for yaskawa robot

A RAG to deal with multiple documents simultaneously using Weaviate, Mistral-7B-v0.1 and Langchain.

Implementing Vector Database on CoNaLa dataset to retrieve program snippets relevant to user queries. This is a very simple simulation of a Vector Database.

This repo contains codes for RAG using docling on colab notebook with langchain, milvus, huggingface embedding model and LLM

👽SFSeeker - an AI assistant with a semantic engine/question writing tutor in Sci-Fi Stack Exchange service

This project uses Python, Hugging Face (sentence-transformers), Milvus + Docker (container running Vector DB) to create a vector database, populate it with details of many people (names, ages, salaries, addresses and their introductions) and enable searching and querying on the database contents using Cosine-Similarity distances on IVF Flat index.

This Multi-Modal Movie Recommendation System leverages a combination of structured numerical features and deep text embeddings to provide accurate and personalized movie recommendations. Unlike traditional recommender systems that rely solely on user ratings or metadata, this model integrates numerical attributes (such as popularity and ratings)

Sistema de recomendação desenvolvido para o G1, da Globo.

all of my master's thesis.

KAIST cs632 Embedded OS Exercise

This project uses Python, Hugging Face (sentence-transformers), Milvus + Docker (container running Vector DB) to create a vector database, populate it with details of many people (names, ages, salaries, addresses and their introductions) and enable searching and querying on the database contents using Cosine-Similarity distances on IVF Flat index.

Streamlit chatbot app that can answer questions based on a pdf. see the readme for more information

This project uses Python, Hugging Face (sentence-transformers), Milvus + Docker (container running Vector DB) to create a vector database, populate it with details of many people (names, ages, salaries, addresses and their introductions) and enable searching and querying on the database contents using Cosine-Similarity distances on IVF Flat index.