cloudcasting-app

This repo is used to run the OCF-ATI cloudcasting model live in production. This model takes previous frames of EUMETSAT satellite images and forecasts the future frames to come.

The repo associated with training the models run here is https://github.com/openclimatefix/sat_pred

The model checkpoints are hosted at: https://huggingface.co/openclimatefix/cloudcasting_uk

Environment Variables

The following environment variables are used in the app:

• SATELLITE_ZARR_PATH: The path to the satellite data in Zarr format.
• OUTPUT_PREDICTION_DIRECTORY: The directory where results are saved.

Optional Environment Variables

• SATELLITE_SCALE_FACTOR: The scale factor for the satellite data. Defaults to 1023.
• SATELLITE_15_ZARR_PATH: The path to the 15 minute satellite data in Zarr format. If this is not set then the SATELLITE_ZARR_PATH is used by .zarr is repalced with _15.zarr

Installation

Setup / Installation

git clone https://github.com/openclimatefix/cloudcasting-app
cd cloudcasting-app
pip install .

Example usage

Running the app locally

It is possbile to run the app locally by setting the required environment variables listed at the top of the app, these should point to the relevant paths for loading satellite data and saving predicitons.

Development

Running the test suite

The test suite is via pytest and can be run from command line using

pytest

Contributing and community

• PR's are welcome! See the Organisation Profile for details on contributing
• Find out about our other projects in the here
• Check out the OCF blog for updates
• Follow OCF on LinkedIn

Contributors

James Fulton 💻

devsjc 🚇

Peter Dudfield 💻

Sumana Sree Angajala 🚇

---

Part of the Open Climate Fix community.