Skip to content

Latest commit

 

History

History
117 lines (66 loc) · 3.61 KB

README.md

File metadata and controls

117 lines (66 loc) · 3.61 KB
Raw

Lecture Chat

Introduction

The Lecture Chat project is designed to offer valuable information and assistance on lectures. It utilizes various resources like lecture slides, screencasts, PDFs, and YouTube links to provide comprehensive answers to user questions. Notably, each response includes citations and links to the relevant source material, ensuring a deeper understanding and access to additional insights.

Requirements

  1. On-Premise Solution: The chatbot operates on an on-premise infrastructure to minimize costs and ensure data privacy.

  2. Docker Containers: Docker containers are employed for easy deployment and management of services.

  3. Open Source Development: The development of the chatbot exclusively utilizes open-source solutions.

Development Resources

LLM Chaining

Langchain Low-Code No-Code

Local Deployment of LLMs on CPU

Python UI

(Vector) Databases

Other Links

Deployment

Follow these steps to run the provided services using Docker Compose:

  1. Install Docker and Docker Compose:

    Ensure that Docker and Docker Compose are installed on your system. If not, you can download and install them from the official Docker website: Docker and Docker Compose.

  2. Clone the Repository:

    Clone this repository to your local machine.

  3. Start Services:

    Start the services defined in the docker-compose.yml file using the following command:

    docker-compose up -d

    The -d option stands for "detach" mode, which runs the services in the background.

  4. Access User Interfaces:

    After starting the services, you can access the various user interfaces using the following URLs:

Accessible User Interfaces (UIs)

  1. Lecture Chat Interface (lecture-chat):

    The Lecture Chat interface is accessible via a web browser by visiting:

    http://localhost:8080

    This interface allows interaction with the Lecture Chat Bot.

  2. Language Processing Tool Interface (langflow):

    The Language Processing Tool interface can be accessed at:

    http://localhost:7860

    Here, the functions of the langflow service for language processing are available.

  3. Code Syntax Highlighting Interface (chroma):

    The interface for the code syntax highlighting service is reachable at:

    http://localhost:8000

    This interface allows testing and using the chroma service's code syntax highlighting functions.

Stopping the Services

To stop and remove the services, execute the following command in the directory containing the docker-compose.yml file:

docker-compose down

This will stop the services and remove the associated containers, networks, and volumes.

License

This project is released under the MIT License.