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
coincident.search.search(dataset='3dep' outputs simplified polygons from simplify_footprint() in src/coincident/search/wesm.py
Hasn't been a major issue except when I was looking at use case workunit == "CO_WestCentral_2019" where the simplified polygon via the 'coincident' search is drastically different than the actual WESM polygon.
Could implement an argument for toggling simplify() and some of it's parameters?
The text was updated successfully, but these errors were encountered:
To speed up the initial search, I actually just created a separate parquet file of convex hulls rather than simplified polygons of all geometries! I wanted to be able to query the remote WESM.gpkg efficiently, but couldn't figure out a way. So opted for a 2 stage search to 1. find workunit/feature id based on the fast convex hull search then 2. use the feature id to just get the single detailed polygon from the remote WESM (which is fast).
in #4 I've switched to just initially searching WESM working bounding boxes . Remote querying the native resolution polygons of the WESM.gpkg is very slow via ogr/geopandas, so the options are really to 1. pre-fetch the entire 1.8GB file or be 2. clever about searching.
Could be good to add a documentation page showing the differences of bbox, convex_hull, and actual polygon.
version: 0.1.dev28+gf5aee55
Python: 3.12.5 custom env
coincident.search.search(dataset='3dep'
outputs simplified polygons fromsimplify_footprint()
in src/coincident/search/wesm.pyHasn't been a major issue except when I was looking at use case
workunit == "CO_WestCentral_2019"
where the simplified polygon via the 'coincident' search is drastically different than the actual WESM polygon.Could implement an argument for toggling simplify() and some of it's parameters?
The text was updated successfully, but these errors were encountered: