Skip to content

A chatbot app using LlamaIndex and GPT-4 with Streamlit

Notifications You must be signed in to change notification settings

maria-lagerholm/chatbot_ai

Repository files navigation

🤖 chatbot_ai

Chatbot Language Framework Model

📚 Overview

A “chatbot_ai” is a Swedish-speaking application designed to provide native Swedish speakers with a fast and informative interaction. The client aimed to offer a platform that fluently understands and responds in Swedish and only answers relevant to construction engineering questions.

Using LlamaIndex and GPT-4 LLM as well as Swedish-trained SpaCy model to make sure that the conversation is relevant to construction engineering and is in Swedish. The application was constructed using Streamlit. Other training information is provided in TXT and CSV files containing companies reports and other relevant documents.

The user-friendly interface was achieved with Lottie animation.

Examaple screenshot

📂 Supported Data Formats

The chatbot supports various data formats for its knowledge base:

  • Text Files (.txt)
  • CSV Files (.csv)
  • URLs

Simply ingest the data, and chatbot_ai will convert it into documents that form the knowledge base.

🔐 Security

Credentials required for the application are securely stored in the secrets.toml file, which is excluded from the version control system using .gitignore

☁️ Deployment

chatbot_ai is deployed on Streamlit Community Cloud.

🔗 Access the Web App

The application is accessible online. Visit chatbot-web.streamlit.app to interact with the ChatBot.

🚀 Quick Start

  1. Clone the repository
  2. Install dependencies
  3. Set up your secrets.toml file with the necessary credentials
  4. Run the Streamlit app

About

A chatbot app using LlamaIndex and GPT-4 with Streamlit

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published