This repo contains code to query PDF(s) using langchain, GPT, chromadb, tiktoken, and streamlit
you can clone the repo and replace the docs with yours and try to run the bot.
For running it locally, perform below steps-
-
git clone https://github.com/codysaint/streamlit-pdf-qa-langchain-app.git
-
cd streamlit-pdf-qa-langchain-app
-
create a new virtual environment called
.venv
python -m venv .venv
-
Activate the virtual environment
.venv\Scripts\activate
-
Install the project requirements
pip install -r requirements.txt
-
Delete the files in
vector_index
directory so as to hold the vectors of your own document. -
Start the app using the following command:
streamlit run app.py