A web application that generates recipes based on ingredients you have on hand!
This is a Node.js Express API server that also serves a React frontned.
The frontend can be found at /index.html (or simply /).
The backend API has a single endpoint. POST /ingredients takes a list of ingredients (as a single string) and then sends a request to OpenAI's chat completion API to generate content around a restaurant dish that can be made with those ingredients.
The implementation uses OpenAI ChatGPT functions, which allows us to specify the JSON structure of the data to be returned, which can more easily be passed as arguments directly to a function in our code (or, in this case, to be passed back to the frontend, which will use the arguments for a function call there).
To use this application, you will need an OpenAI API Key to make calls to the chat completion API.
Note
Every OpenAI request incurs costs, the price may vary depending on the selected model, for example, using the GPT-3 model, in order to reach $1 you'll need to request more than 30,000 times. Please refer to the OpenAI API pricing page for more information.
This project is set to use Node v20.10.0. Install and use this version with volta.
- Install dependencies
yarn install- Make local copy of
.envand customize
cp .env.template .env- Edit
.envto add environment variables
Add OPENAI_API_KEY (which you will need to obstain from your OpenAI account).
If you want to use a port other than the default (3000), add PORT environment variable.
If you want to use a different ChatGPT model than the default (gpt-3.5-turbo-1106), add CHATGPT_MODEL environment variable.
Start API server on (default) port 3000
yarn build && yarn startyarn testCreate an Heroku application:
heroku create <app-name>Be sure to add OPENAI_API_KEY environment variable:
heroku config:set OPENAI_API_KEY=sk-vxxxxxxxAdd the code to your local git repository and deploy to Heroku with:
git push heroku main