Skip to content

Python code to produce validation statistics for a national or regional GB PV forecast.

Notifications You must be signed in to change notification settings

SheffieldSolar/PV-Forecast-Validation

Repository files navigation

PV-Forecast-Validation

Python code to produce validation statistics for a national or regional GB PV forecast.

How does it work?

The code is designed to be run inside a Docker container and will launch a local Flask development server on your machine. You can then visit http://127.0.0.1:5000 and use the web interface to upload some historical PV forecast data. The code will fetch the corresponding "actual" data from the PV_Live API and calculate various error metrics, before presenting them as interactive graphs in your browser.

How do I get set up?

How do I get started?

In your local terminal/command-prompt, run the following command:

docker run -it --rm -p 5000:5000 sheffieldsolar/pv_forecast_validation:latest

Visit http://127.0.0.1:5000 in your browser - that's it!

Exporting to PDF

The graphical reports can be exported to PDF using your browser built-in "print-to-PDF" function. This functionality has only been tested in Chrome browser, other browsers are likely to mess up the layout!

An example report, giving 2019 validation stats for Sheffield Solar's PV_Forecast service, is available here.

About

Python code to produce validation statistics for a national or regional GB PV forecast.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published