This project highlights how to leverage a ChromaDB vectorstore in a Langchain pipeline to create drumroll please a GPT Investment Banker (ergh, I cringed as I wrote that, but alas it's actually pretty practical). You can load in a pdf based document and use it alongside an LLM without the need for fine tuning.
- Create a virtual environment
python -m venv langchainenv
- Activate it:
- Windows:
.\langchainenv\Scripts\activate
- Mac:
source langchain/bin/activate
- Windows:
- Clone this repo
git clone https://github.com/nicknochnack/LangchainDocuments
- Go into the directory
cd LangchainDocuments
- Install the required dependencies
pip install -r requirements.txt
- Add your OpenAI APIKey to line 22 of
app.py
- Start the app
streamlit run app.py
- Go back to my YouTube channel and like and subscribe 😉...no seriously...please! lol
The main LG Agent used:Langchain VectorStore Agents
👨🏾💻 Author: Nick Renotte
📅 Version: 1.?
📜 License: This project is licensed under the MIT License