This service powers the Sol Mate GPT. It uses Modal because it's one of the easiest ways to run a simple service like this.
The output has been specifically formatted in a way that helps GPT visually describe the weather to Dall·E. This output can probably be used to describe the weather for any other LLM-related purpose.
Make sure you have Modal installed:
pip install modal
Then, install the dependencies:
pip install ephem pytz requests
(You may use a virtual environment for this, up to you.)
For now you'll need to update blixt
in the code to your own username.
And finally you can run it to see that it works:
$ modal run sol_mate.py
✓ Initialized. View run at https://modal.com/blixt/apps/ap-NbqYUrhnWHuqB8itLCIcGE
✓ Created objects.
├── 🔨 Created mount /Users/blixt/src/sol-mate/sol_mate.py
└── 🔨 Created weather_api => https://blixt--sol-mate-weather-api-dev.modal.run
The weather for {'latitude': 40.7128, 'longitude': -74.006, 'timezone': 'America/New_York', 'temperature_unit': 'fahrenheit'}:
Temperature: 46°F
Weather condition: overcast
The local time is 13:06 on a Thursday in April.
The sun is hidden behind clouds.
The sky is completely overcast.
A moderate breeze is blowing.
Stopping app - local entrypoint completed.
GET /current -> 200 OK (duration: 4.27 s, execution: 418.0 ms)
✓ App completed. View run at https://modal.com/blixt/apps/ap-NbqYUrhnWHuqB8itLCIcGE