You will need access to the project over on the Google Cloud Platform, and a key for the service account in json format. To get this, head to https://console.cloud.google.com, navigate to the project folder and then go to APIs -> Credentials -> Create credentials -> Service account key. Select 'ml-app' and save the key locally.

Next install direnv, for automatically loading the environment when you navigate to the project folder.

brew install direnv

Follow the setup instructions over at https://direnv.net/ to configure direnv to work with your shell. Once this is done create a .envrc file, copying over from .envrc.sample and adding the necessary information. Run direnv allow to give direnv permission to load your environment.

Finally you can install dependencies:

pip install -r requirements.txt

To run the app:

python3 app.py