New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Let CloughTocher2DInterpolator fit "nearest" for points outside the convex hull (instead of nans) #14386
Comments
The example is contributed by Ajay Shanker Tripathi in scipygh-14386
The example is contributed by Ajay Shanker Tripathi in scipygh-14386
The example is contributed by Ajay Shanker Tripathi in scipygh-14386
* DOC: interpolate: add an example for interp1d(..., fill_value=(left, right)) closes gh-16338 * DOC: interpolate: add an example of CubicSpline.extend to control the extrapolation The example is contributed by Shamus Husheer in gh-14472. Co-authored-by: Shamus Husheer <s.husheer@gmail.com> * DOC: interpolate: add a worked example of interpolation+asymptotics * DOC: interpolate: add an example of CT + nearest-neighbor extrapolation The example is contributed by Ajay Shanker Tripathi in gh-14386
@ev-br thanks so much! Sorry for not replying sooner. If it's not too late, I wouldn't mind being credited. My email is chausies @ stanford.edu . |
Well, the relevant scipy commit does list your name, which was taken from your GH profile: 61ff61f For adding a git-recognized Let me ask our release manager @tylerjereddy if it is still possible to manually add a name to the list of authors for the upcoming 1.10 release? |
@ev-br I manually added @chausies to the author list in #17696 (comment) |
Thank you Tyler! |
Thank you so much @tylerjereddy @ev-br! Now I can tell everyone I've hit the big leagues ^_^. |
Currently, CloughTocher2DInterpolator can only fill in a constant value (nan by default) when queried outside the convex hull of the input points. But this absolutely isn't ideal.
What would be much better is to use the value of the nearest neighbor, which NearestNDInterpolator does by default. For CloughTocher2DInterpolator, which already constructs the Delaunay Triangulation of the input points, finding the nearest point on the convex hull should be efficient.
Here's just my naive implementation for filling in the nearest neighbor value outside the convex hull of the inputs.
The text was updated successfully, but these errors were encountered: