Unable to use factors in raster::predict

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

Unable to use factors in raster::predict

R-sig-geo mailing list
Dear list,

I would like to use the 'predict' function in the 'raster' package in an implementation of species distribution modelling with a couple of factor variables. My case can be set up exactly as the cforest example listed in the help file. Unfortunately, I cannot get the example to work: it throws an error because of the unused argument 'factors', which I need. Here is the code I am referring to:

# create a RasterStack or RasterBrick with with a set of predictor layers
logo <- brick(system.file("external/rlogo.grd", package="raster"))
names(logo)

# known presence and absence points
p <- matrix(c(48, 48, 48, 53, 50, 46, 54, 70, 84, 85, 74, 84, 95, 85,
              66, 42, 26, 4, 19, 17, 7, 14, 26, 29, 39, 45, 51, 56, 46, 38, 31,
              22, 34, 60, 70, 73, 63, 46, 43, 28), ncol=2)

a <- matrix(c(22, 33, 64, 85, 92, 94, 59, 27, 30, 64, 60, 33, 31, 9,
              99, 67, 15, 5, 4, 30, 8, 37, 42, 27, 19, 69, 60, 73, 3, 5, 21,
              37, 52, 70, 74, 9, 13, 4, 17, 47), ncol=2)

# extract values for points
xy <- rbind(cbind(1, p), cbind(0, a))
v <- data.frame(cbind(pa=xy[,1], extract(logo, xy[,2:3])))

# cforest (other Random Forest implementation) example with factors argument

v$red <- as.factor(round(v$red/100))
logo$red <- round(logo[[1]]/100)

library(party)
m <- cforest(pa~., control=cforest_unbiased(mtry=3), data=v)
f <- list(levels(v$red))
names(f) <- 'red'
pc <- predict( m,logo, OOB=TRUE, factors=f)

Error in RET@prediction_weights(newdata = newdata, mincriterion = mincriterion,  :
  unused argument (factors = f)


Note that I needed to change the order in the arguments in the last line. If I run it the way it was in the example, like this:

pc <- predict(logo, m, OOB=TRUE, factors=f)


then the error is the following:


Error in v[cells, ] <- predv :

  number of items to replace is not a multiple of replacement length

Also note that if I run the line without the 'factors' argument the resulting value (r) is a matrix, and not a raster object.

I am using Package 'raster' version 2.6-7 on Windows 7 (where this worked fine in the past).

Many thanks for any help,

Genoveva

Genoveva Gonzalez Mirelis, Scientist
Institute of Marine Research
Nordnesgaten 50
5005 Bergen, Norway
Phone number +47 55238510


        [[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: Unable to use factors in raster::predict

Hugo Costa
This is not of much help, but I run the code, and the line
pc <- predict(logo, m, OOB=TRUE, factors=f)
did work fine. I got a Rasterlayer, which is the R logo with pixels ranging
from 0 to 1.
Hugo

Gonzalez-Mirelis, Genoveva via R-sig-Geo <[hidden email]> escreveu
no dia terça, 20/11/2018 à(s) 10:23:

> Dear list,
>
> I would like to use the 'predict' function in the 'raster' package in an
> implementation of species distribution modelling with a couple of factor
> variables. My case can be set up exactly as the cforest example listed in
> the help file. Unfortunately, I cannot get the example to work: it throws
> an error because of the unused argument 'factors', which I need. Here is
> the code I am referring to:
>
> # create a RasterStack or RasterBrick with with a set of predictor layers
> logo <- brick(system.file("external/rlogo.grd", package="raster"))
> names(logo)
>
> # known presence and absence points
> p <- matrix(c(48, 48, 48, 53, 50, 46, 54, 70, 84, 85, 74, 84, 95, 85,
>               66, 42, 26, 4, 19, 17, 7, 14, 26, 29, 39, 45, 51, 56, 46,
> 38, 31,
>               22, 34, 60, 70, 73, 63, 46, 43, 28), ncol=2)
>
> a <- matrix(c(22, 33, 64, 85, 92, 94, 59, 27, 30, 64, 60, 33, 31, 9,
>               99, 67, 15, 5, 4, 30, 8, 37, 42, 27, 19, 69, 60, 73, 3, 5,
> 21,
>               37, 52, 70, 74, 9, 13, 4, 17, 47), ncol=2)
>
> # extract values for points
> xy <- rbind(cbind(1, p), cbind(0, a))
> v <- data.frame(cbind(pa=xy[,1], extract(logo, xy[,2:3])))
>
> # cforest (other Random Forest implementation) example with factors
> argument
>
> v$red <- as.factor(round(v$red/100))
> logo$red <- round(logo[[1]]/100)
>
> library(party)
> m <- cforest(pa~., control=cforest_unbiased(mtry=3), data=v)
> f <- list(levels(v$red))
> names(f) <- 'red'
> pc <- predict( m,logo, OOB=TRUE, factors=f)
>
> Error in RET@prediction_weights(newdata = newdata, mincriterion =
> mincriterion,  :
>   unused argument (factors = f)
>
>
> Note that I needed to change the order in the arguments in the last line.
> If I run it the way it was in the example, like this:
>
> pc <- predict(logo, m, OOB=TRUE, factors=f)
>
>
> then the error is the following:
>
>
> Error in v[cells, ] <- predv :
>
>   number of items to replace is not a multiple of replacement length
>
> Also note that if I run the line without the 'factors' argument the
> resulting value (r) is a matrix, and not a raster object.
>
> I am using Package 'raster' version 2.6-7 on Windows 7 (where this worked
> fine in the past).
>
> Many thanks for any help,
>
> Genoveva
>
> Genoveva Gonzalez Mirelis, Scientist
> Institute of Marine Research
> Nordnesgaten 50
> 5005 Bergen, Norway
> Phone number +47 55238510
>
>
>         [[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