Faster subset gridpoint #452
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Pull Request Checklist:
What kind of change does this PR introduce?:
When subsetting a curvilinear dataset (2D lat lon) to gridpoints, use a
scipy.spatial.KDTreeto find nearest neighbours with euclidean distance in lat/lon space, instead of computing the great circle distance (in meters) for all points.We are already using a lat/lon euclidean distance for the rectilinear case (1D lat / lon). The loss in precision is compensated by a significant performance boost. For example, my use case needed to extract 94 points from a 800x1000 grid. Instead of computing 94x1000x800 great circle distances, we now only need to compute 94 when tolerance is passed. None otherwise.
Before this change, my use case took ~120 s and now it takes 350 ms.
I also modified how we get the lat and lon coordinate to use the utils instead of relying on
latandlonnames. And I modified these utils so a variable named "lon" is detected as a longitude (and similarly for "lat").Does this PR introduce a breaking change?:
Kinda as the distance metric has changed. In most cases, I don't expect different result, but there could be some extreme cases with points near the poles where a different neighbour is now choosed.
Other information: