Predict expenditure changes using OpenAI API using forecasted CPI.

backend

uses flask to stand-up endpoints that can take parameters and hit openai api with predefined prompts

required:

• python
• poetry

cd into backend folder

set local env var for openai key:

export openai_api_key=blah

install dependencies:

poetry install

run app:

poetry run flask run

endpoint to see the current cpi:

http://localhost:5000/api/cpi/value

endpoint to hit openai:

http://localhost:5000/api/cpi/process

values will be passed through from the frontend

run linter:

poetry run black .

frontend

uses vite to stand-up a react web ui that takes input from users, which are passed to the backend to hit openai with a prompt and display the response

required:

• yarn

cd into frontend folder

install dependencies:

yarn install

run app:

run yarn dev

access web ui:

http://localhost:5173/

run linter:

yarn lint