You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It'd be great to show some of the FTW model 'output' on the map as vectors.
I believe ASU is running the models in 16 different places, so if they can put the raster or vector (ftw inference polygonize output) somewhere (could do a new repo on source cooperative) then me or @m-mohr can pretty easily convert to pmtiles and get on the map.
We should style it differently, and probably make it an independent layer that can be turned on/off.
This would make it even nicer to have a satellite background, to see the results. I think it'd be fine if it was just when zoomed in - then just make like a planetary computer STAC or sentinel hub request. We could bake the s2 scene needed into the output, but as long as there's an API to get the scenes then determination_datetime and geographic is likely sufficient?
The text was updated successfully, but these errors were encountered:
It'd be great to show some of the FTW model 'output' on the map as vectors.
I believe ASU is running the models in 16 different places, so if they can put the raster or vector (
ftw inference polygonize
output) somewhere (could do a new repo on source cooperative) then me or @m-mohr can pretty easily convert to pmtiles and get on the map.We should style it differently, and probably make it an independent layer that can be turned on/off.
This would make it even nicer to have a satellite background, to see the results. I think it'd be fine if it was just when zoomed in - then just make like a planetary computer STAC or sentinel hub request. We could bake the s2 scene needed into the output, but as long as there's an API to get the scenes then determination_datetime and geographic is likely sufficient?
The text was updated successfully, but these errors were encountered: