I am working with GRASS and I am trying to do a supervised classification. The issue here is that I get an error for NaN values.

ERROR: Not enough data points

Now I know that is an error that you get, most of the times, when the projection is different and because I know that I re-projected all of my data in GGRS87 (I have both vector and raster data) but I still get the same error. What should I do? Should I try another projection or is there another way to do supervised classification.

P.S I tried with this tutorial https://grasswiki.osgeo.org/wiki/Image_classification#Supervised_classification

and I tried to create a raster map from CORINE2000

g.region vect=corine align=raster v.to.rast in=corine out=corine_training use=attr col=CODE labelcol=CODE2 --o

And now I can't generate statistics using

i.gensigset group=OB subgroup=raster_maps sig=corine_smap training=corine_training

1 Answer 1


So researching for my issue I came across an old github repository filled with lectures on how to do a Supervised Classification


Which actually helped me retrace my steps and solve my problem.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.