Streamlit web app providing a Large Language Model (LLM) chatbot with a custom knowledge base
- Deployed to Streamlit Community Cloud ☁️
- Conversation LLM chatbot chain only
- For simplicity and privacy, the semantic search of personal or organizational documents to create a custom knowledge base for the LLM is available only as a local deployment feature
- URL: https://ask-chatbot.streamlit.app
- Extend the LLM chatbot with a custom knowledge base
- Include the data you wish to upload to extend the LLM chatbot in the 'data' directory
- Anticipates to load any .txt, .pdf, .csv, .docx, or .xlsx files found in the directory
- Web scraping, see code for details
- Data is persisted within a local ChromaDB vector store
- Developed with Python version 3.10
- Streamlit
- LangChain
- OpenAI
- ChromaDB
- BeautifulSoup
-
Create a .streamlit folder in the 'src' directory and include the contents of the .streamlit.example
-
To initialize a virtual enviroment, navigate to the 'src' directory in the terminal and execute
$ python -m venv venv -
Activate the virtual environment, on MacOS or Linux with
$ source venv/bin/activate -
Install dependencies
$ pip install -r requirements.txtSome LangChain packages may have hidden dependencies that require pip installation
-
Run the app
$ streamlit run app.pyOpen your browser at http://localhost:8501
-
To deactivate the virtual environment
$ deactivate
| LLM Conversation |
|---|
 |
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent