LangChain Project

This project is a collection of various applications and scripts utilizing the LangChain library for natural language processing tasks.

Project Structure

.env .gitignore agents/ - agents.ipynb api/ - app.py - client.py chatbot/ - app_llama.py - app.py rag/ - attention.pdf - chain.ipynb - rag.ipynb - speech.txt requirements.txt

Folders and Files

• agents/: Contains Jupyter notebooks for agent-based tasks.

  • agents.ipynb: Notebook for agent-related experiments.

• api/: Contains the API implementation using FastAPI.

  • app.py: Main API application file. Implements routes and initializes models.
  • client.py: Client for interacting with the API.

• chatbot/: Contains applications for chatbot implementations.

  • app_llama.py: Chatbot application using the LLaMA model.
  • app.py: Main chatbot application using Streamlit.

• rag/: Contains resources and notebooks for retrieval-augmented generation (RAG) tasks.

  • attention.pdf: PDF document related to attention mechanisms.
  • chain.ipynb: Notebook for chain-based RAG tasks.
  • rag.ipynb: Notebook for RAG experiments.
  • speech.txt: Text file related to speech processing.

• .env: Environment variables file.

• .gitignore: Git ignore file.

• requirements.txt: Python dependencies file.

Setup

1. Clone the repository:

   git clone <repository-url>
   cd LangChain

2. Create and activate a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`

3. Install the dependencies:

   pip install -r requirements.txt

4. Set up environment variables:

   • Copy the .env file to the root directory and update the values as needed.

Running the Applications

API

To run the FastAPI application:

cd api
uvicorn app:app --reload

Chatbot

To run the Streamlit chatbot application:

cd chatbot
streamlit run app.py

Usage

API: Access the API endpoints at http://localhost:8000.
Chatbot: Open the Streamlit application in your browser and interact with the chatbot.