Krzysztof Dyba

Here is @appelmar great presentation and examples about this topic: https://appelmar.github.io/CONAE_2022/

Since the first post on this topic, the situation has improved and more training materials and examples have appeared. Here is the list of interesting tutorials on processing large Earth...

CRAN Spatial Task View is maintained by our friends Roger and Jakub, so don't worry ;) I think this is good idea because some people (including me) use these lists...

> What do you think? Always build the tree or optionally or have separate functions for it? Maybe optionally? In `{sf}` there is the `prepared` argument. In `{terra}` it works...

@Robinlovelace, little offtopic, but I tried `{sfheaders}` to extract coordinates from geometry and it is slower than `{sf}`. However, it doesn't really matter because this operation has little overhead. The...

BTW @Robinlovelace, isn't the `world` object CRS WGS84? Then you are comparing the `s2` spherical engine in `{sf}` with planar in `{geos}` and `{rsgeo}`. From my observations, in general `GEOS`...

I adapted your example from the [vignette](https://jakubnowosad.com/supercells/articles/rgb_vars.html) and here are my results with code. Basically, I used kmeans with coordinates and pixel smoothing. If we omit the coordinates scaling, then...

The following questions arise: 1. Does any method create better superpixels / clusters (what are the differences)? 2. What is the performance of these methods? 3. Can they be applied...

> Could you also try to prepare a larger example (e.g, 10000 by 10000 cells)? Do you have such dataset? If not then we can use the Sentinel 2 image...

Some notes: 1. I think that for larger datasets we should train models on smaller sample and then predict on the whole dataset. 2. kmeans algorithm is probably not the...