/llm-builder-VertaAI

Primary LanguagePython

Document Summarization Bot

The quickstart example is a document summarization bot that can take a file, URL or text as input and summarize the content. It uses Open AI APIs. It also includes a LLM builder tool to help you manage and iterate on prompts and input datasets. A set of sample prompts and datasets are already available and you can bring your own data and refine on prompts.

Setup steps

  • Python 3 in your machine

  • Setup and activate python virtual environment

    • python3 -m venv yourownenv
    • source yourownenv/bin/activate

  • Download or git-clone the quickstart project

  • Go to main folder of the project

  • Sign-up in Verta and get your DevKey - https://app.verta.ai/sign-up

    • Run export VERTA_EMAIL=<your verta email>
    • Run export VERTA_DEV_KEY=<your verta devkey>

  • You can either use your own Open AI DevKey or Verta will provide a temporary key Run export OPENAI_API_KEY=<openAI key>

  • Run pip install -r requirements.txt to install all of the project dependencies.

  • Go to src folder and run streamlit run LLM_Builder.py to start the app.

  • The quickstart app will open in localhost. If it doesn't, you can copy and paste the URL that was printed in the terminal.