https://github.com/runtime-error786/flash-reranker
https://github.com/runtime-error786/flash-reranker
huggingface langchain llama3
Last synced: 10 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/runtime-error786/flash-reranker
- Owner: runtime-error786
- Created: 2024-08-22T12:36:22.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-08-22T13:45:56.000Z (over 1 year ago)
- Last Synced: 2025-02-01T08:22:18.600Z (12 months ago)
- Topics: huggingface, langchain, llama3
- Language: Jupyter Notebook
- Homepage:
- Size: 6.2 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# LangChain Document Retrieval and QA System
This project demonstrates a document retrieval and question-answering system using LangChain. It integrates document loading, text splitting, embeddings, vector storage, and QA retrieval with advanced LLMs. This setup utilizes pre-built components from LangChain and other libraries for a streamlined information retrieval and response system.
## Features
- **Document Loading**: Loads PDF documents and extracts text.
- **Text Splitting**: Splits large documents into manageable chunks.
- **Embedding**: Converts text chunks into vector embeddings.
- **Vector Storage**: Stores and retrieves vector embeddings using Chroma.
- **Contextual Compression Retrieval**: Uses `FlashrankRerank` for document reranking.
- **QA System**: Uses `RetrievalQA` for generating answers from documents.