Neura Market
Langchain Semantic Search logo

Langchain Semantic Search

Free

Search and indexing your own Google Drive Files using GPT3, LangChain, and Python

FreeFree tier
Type
Open Source
Visit Website GitHub (40 stars)

About Langchain Semantic Search

Langchain Semantic Search is an open-source Jupyter notebook tool that allows users to index and semantically search their Google Drive files using GPT-3, LangChain, and FAISS. It provides a workflow to build a FAISS vector index over a corpus of documents and then query it with natural language questions, returning relevant answers. The repository includes a sample dataset of neuromodulation papers for testing. Users must copy the sample folder to their own Google Drive, set the path, and run the notebook to perform semantic searches. The tool leverages LangChain's integration with GPT-3 for generating embeddings and answering queries, making it suitable for personal document retrieval.

Key Features

Indexes Google Drive files using FAISS for fast similarity search
Uses GPT-3 via LangChain for semantic understanding and answer generation
Provides a Jupyter notebook with step-by-step workflow
Supports custom document corpora (sample neuromodulation papers provided)
Enables natural language queries against indexed documents

Pros & Cons

Pros
  • Leverages powerful GPT-3 model for accurate semantic search and answer generation
  • Open-source and free to use (no subscription fees for the tool itself)
  • Easy to set up with a provided Jupyter notebook and sample data
  • Customizable to any Google Drive folder containing compatible document types
Cons
  • Requires setup of Google Drive API and authentication
  • GPT-3 usage incurs API costs (not included in the tool)
  • Limited to Jupyter notebook environment; no web interface or standalone app
  • Indexing speed may be slow for large document collections (author notes planned improvements)
  • Dependent on LangChain and FAISS library installations

Best For

Searching personal or research document collections stored in Google DriveQuickly retrieving relevant information from a large corpus of PDFs or text filesBuilding a semantic search engine for academic papers or project documentationExploring GPT-3's ability to answer questions based on a private knowledge base

FAQ

How do I get started with this tool?
Clone the repository, run the Jupyter notebook 'langchain_semantic_search.ipynb', and follow the instructions to set your Google Drive path and query questions.
What files can I index from Google Drive?
The tool is designed to work with document files (likely text, PDFs, etc.) that LangChain can process. The sample dataset consists of neuromodulation papers.
Is this tool free?
The tool itself is open-source and free. However, you will need your own OpenAI API key, and GPT-3 usage is billed by OpenAI.