Random forest script - problems with training areas polygon?

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Random forest script - problems with training areas polygon?

Silvia Flaherty
Hi all,



I have been working with the random forest R script and have had no
problems at all up to now.

I recently modified the training areas shapefile. It´s actually an edited
version of the original shapefile (which worked just fine). Same
projection, etc, only the areas are smaller and more specific. That's the
only input I changes. The script reads the shapefile but it doesn't do the
classification.



The error  I get is:

 Error in `[.data.frame`(satValues, , 2) : undefined columns selected



Checking line by line, he problem seems to be (line 139 of the loop):

satValues <- extract(satImage, class_data[i,], cellnumbers=TRUE, df=TRUE)





I´m not very skilled at using R but I´m quite OK with GIS. I have tried
editing the shapefile again, saving it as another file, etc. Same result.



Any ideas what could be going wrong?

Thanks!



Silvia


<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Libre
de virus. www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
<#m_542342069031620777_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: Random forest script - problems with training areas polygon?

Roman Luštrik
Can you make a reproducible example (data, scripts)? Have you tried
debugging your code?

Cheers,
Roman

On Wed, Nov 7, 2018 at 2:36 PM Silvia Flaherty <[hidden email]>
wrote:

> Hi all,
>
>
>
> I have been working with the random forest R script and have had no
> problems at all up to now.
>
> I recently modified the training areas shapefile. It´s actually an edited
> version of the original shapefile (which worked just fine). Same
> projection, etc, only the areas are smaller and more specific. That's the
> only input I changes. The script reads the shapefile but it doesn't do the
> classification.
>
>
>
> The error  I get is:
>
>  Error in `[.data.frame`(satValues, , 2) : undefined columns selected
>
>
>
> Checking line by line, he problem seems to be (line 139 of the loop):
>
> satValues <- extract(satImage, class_data[i,], cellnumbers=TRUE, df=TRUE)
>
>
>
>
>
> I´m not very skilled at using R but I´m quite OK with GIS. I have tried
> editing the shapefile again, saving it as another file, etc. Same result.
>
>
>
> Any ideas what could be going wrong?
>
> Thanks!
>
>
>
> Silvia
>
>
> <
> https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail
> >
> Libre
> de virus. www.avast.com
> <
> https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail
> >
> <#m_542342069031620777_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> [hidden email]
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>


--
In God we trust, all others bring data.

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: Random forest script - problems with training areas polygon?

Hugo Costa
Silvia,
you say that the areas of the shapefile are smaller. Do they intersect
pixels centroids? The help of function 'extract' says the following (which
might be relevant for your case?):

If y represents polygons, the extract method returns the values of the
cells of a Raster* object that are covered by a polygon. A cell is covered
if its center is inside the polygon (but see the weights  option for
considering partly covered cells; and argument small for getting values for
small polygons anyway).

Cheers
Hugo

Roman Luštrik <[hidden email]> escreveu no dia quarta, 7/11/2018
à(s) 15:03:

> Can you make a reproducible example (data, scripts)? Have you tried
> debugging your code?
>
> Cheers,
> Roman
>
> On Wed, Nov 7, 2018 at 2:36 PM Silvia Flaherty <[hidden email]>
> wrote:
>
> > Hi all,
> >
> >
> >
> > I have been working with the random forest R script and have had no
> > problems at all up to now.
> >
> > I recently modified the training areas shapefile. It´s actually an edited
> > version of the original shapefile (which worked just fine). Same
> > projection, etc, only the areas are smaller and more specific. That's the
> > only input I changes. The script reads the shapefile but it doesn't do
> the
> > classification.
> >
> >
> >
> > The error  I get is:
> >
> >  Error in `[.data.frame`(satValues, , 2) : undefined columns selected
> >
> >
> >
> > Checking line by line, he problem seems to be (line 139 of the loop):
> >
> > satValues <- extract(satImage, class_data[i,], cellnumbers=TRUE, df=TRUE)
> >
> >
> >
> >
> >
> > I´m not very skilled at using R but I´m quite OK with GIS. I have tried
> > editing the shapefile again, saving it as another file, etc. Same result.
> >
> >
> >
> > Any ideas what could be going wrong?
> >
> > Thanks!
> >
> >
> >
> > Silvia
> >
> >
> > <
> >
> https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail
> > >
> > Libre
> > de virus. www.avast.com
> > <
> >
> https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail
> > >
> > <#m_542342069031620777_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
> >
> >         [[alternative HTML version deleted]]
> >
> > _______________________________________________
> > R-sig-Geo mailing list
> > [hidden email]
> > https://stat.ethz.ch/mailman/listinfo/r-sig-geo
> >
>
>
> --
> In God we trust, all others bring data.
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> [hidden email]
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo