Skip to content
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

Getting kriging interpolated grid with None values where the samples are far from the semivariogram range #134

Closed
carlosfelg opened this issue Jan 13, 2020 · 9 comments
Assignees
Labels

Comments

@carlosfelg
Copy link

How can I get an interpolated grid with None (or no data) values where all the data neighbours (samples) are far from the variogram range value? I mean, I would like to use the variogram range value as constraining when there is no input data inside the range radio.

@MuellerSeb
Copy link
Member

What do you mean by "variogram range value"? I really don't understand what you mean.
To try a shot in the dark, one thing of interest could be the estimated mean for ordinary krging. Do you mean that?

@carlosfelg
Copy link
Author

carlosfelg commented Jan 27, 2020 via email

@MuellerSeb
Copy link
Member

In OrdinaryKriging.execute

def execute(self, style, xpoints, ypoints, mask=None, backend='vectorized',

you can set n_closest_points to only take a certain amount of nearby values into account. Maybe this is helps?

@MuellerSeb MuellerSeb reopened this Jan 28, 2020
@MuellerSeb MuellerSeb self-assigned this Jan 28, 2020
@carlosfelg
Copy link
Author

What distance value is considered, inside the ordinary kriging function, to define the closest points? The n_closest_points parameter is the minimum or the maximum number of closest points?

@carlosfelg
Copy link
Author

carlosfelg commented Jan 28, 2020 via email

@MuellerSeb
Copy link
Member

n_closest_points is exact unless there are less data points in total. The n clostest points are determined by scipys KDTree

...but it seems that it does not work properly.

What exactly do you mean?

@carlosfelg
Copy link
Author

carlosfelg commented Jan 29, 2020 via email

@MuellerSeb
Copy link
Member

MuellerSeb commented Jan 29, 2020

The number of the closest points is fixed and not determined by a search radius. If you have a dense data set, the n closest points would be rather near.

This needs to be improved in the future.

@MuellerSeb
Copy link
Member

#143 addresses the missing search radius.

Closing for now due to inactivity. Feel free to re-open.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

2 participants