CUSF predictor running locally?

153 views
Skip to first unread message

Andrew Mulholland

unread,
Apr 13, 2021, 1:01:34 PM4/13/21
to UKHAS
I was looking to run a copy of the CUSF predictor locally to play around with the hourly predictor. There are plenty of forks of it though it seems across Github with most not touched in years. Anyone able to point me towards the one actually used by http://predict.habhub.org?

Mark Jessop

unread,
Apr 13, 2021, 5:42:37 PM4/13/21
to uk...@googlegroups.com
The one on predict.habhub.org is using a forked version of the cusf-standalone-predictor to use the full wind model files downloaded by https://github.com/cuspaceflight/tawhiri-downloader
As you mention, the actual code that's running on the site is buried in a branch somewhere that I can't find anymore.. 

A better option nowadays would be to make something that uses the Tawhiri API: https://tawhiri.readthedocs.io/en/latest/api.html
Tawhiri is much faster than the older predictor. With this, the predictions could be polled in the client (browser) instead of being run in batches on the servers.

I had a go at porting the current habhub predictor page to Leaflet, as well as to using the Tawhiri API. That's live here: https://predict.sondehub.org/ and the code is here: https://github.com/darksidelemm/leaflet_predictor

73
Mark VK5QI

On Wed, Apr 14, 2021 at 2:31 AM Andrew Mulholland <gba...@gmail.com> wrote:
I was looking to run a copy of the CUSF predictor locally to play around with the hourly predictor. There are plenty of forks of it though it seems across Github with most not touched in years. Anyone able to point me towards the one actually used by http://predict.habhub.org?

--
You received this message because you are subscribed to the Google Groups "UKHAS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ukhas+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ukhas/32a3683d-4b31-484e-a4ed-913c66947e30n%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages