Simple inverse distance weighted (IDW) interpolation with python

simple_idw.ipynb

Thanks for sharing this code.

I think in terms of visualizing the grid, it works better, at least for points projected in the UTM prj:

plt.imshow(
    grid,
    extent=(x.min(), x.max(), y.min(), y.max()),
    cmap='rainbow',
    interpolation='gaussian',
    origin="lower")

can I pease ask that how I can import data from a real dataset set file foe example excel forman?

can i know that how to save this output Raster format?

@TuhinaN can you specify what you mean/what?

to convert the graphics to raster and save it to pdf:
https://www.tutorialspoint.com/how-to-save-figures-to-pdf-as-raster-images-in-matplotlib

@Majramos how can measure the accuracy of the different power values that I am using (like 1 or 2 or 5), also, is it possible to run the power values in a loop, like one out put for power 1, one for 2 and so on... and then try to understand which power parameters provide best interpolation.

Thank you for the code ! I am not really sure of this but I think that if you want to change the norm you can't use the hypothenus anymore.
distance = (np.abs(d0)**p + np.abs(d1)**p)**(1/p)
would be for the Lp-norm in general

Now to try and figure out how to avoid memory issues on 1000x1000 image :)