RpiGPT

(Name to be changed) Visualization of trajectories and correlations with self-report surveys with speech and language use from users.

  1. Download data and put it in empathic_stories/
  2. python encode_data.py creates a vector and text db of the stories for a station
  3. python run_agent.py runs a LLM agent on the db for queries

Open two terminals

  1. for the server (local)
  2. for the client (local)

Running the Server - Flask

In one of the terminals run the server

create a virtual environment:

python -m venv venv

Activate the virtual environment:

On Windows (PowerShell):

.\venv\Scripts\Activate.ps1

On Unix or MacOS:

source venv/bin/activate

Next, install the required Python packages:

pip install -r requirements.txt

Finally, run the server

python server.py