Skip to content

Latest commit

 

History

History
38 lines (22 loc) · 802 Bytes

File metadata and controls

38 lines (22 loc) · 802 Bytes

Generating visualizations with Gen AI and D3.js

This repo contains an example of structured data extraction using a multimodal LLM, and its visualization with a simple D3.js code.

Clone the repo

git clone git@github.com:recursiveai/sust_hackathon_2024.git

Setup

Install all of the dependencies

npm install

Setup a python venv and install the requirements, load your GEMINI_KEY from the .env file and run

pip install -r requirements.txt
source .venv/bin/activate
source .env

Extract data

python process_data.py data/table.png data/forest.json

Visualize data

Start a simple server and see the result in localhost:5713

npm run dev

Running tests

npm run test

This will start a watch on all of your files and rerun tests that on the affected codepath.