plot.predict.gradientForest: Plot a 'gradientForest' prediction in geographical space

In gradientForest: Random Forest functions for the Census of Marine Life synthesis project.

Plot a gradientForest prediction in geographical space

Description

Plot method for predict.gradientForest and predict.combinedGradientForest objects. A colour representation of the biologically transformed space is plotted on geographical space.

Usage

## S3 method for class 'predict.gradientForest'
plot(x, X, Y, palette = "rg", col3 = 127, ...)

Arguments

x	an object of class predict.gradientForest generated by predict.gradientForest or predict.combinedGradientForest
X	vector of positions in the X direction, having same length as nrow(x)
Y	vector of positions in the X direction, having same length as nrow(x)
palette	controls the colour key of the map.
col3	controls the amount (0-255) of the third colour in the palette. By default there is no blue.
...	further arguments passed to or from other methods.

Details

The first and second principal components of x are plotted on X and Y as a colour, with, by default (palette="rg"), the red component proportional to the first PC and green proportional to the second PC. Other choices of palette, are: "gr", "br", "rb", "bg" and "gb".

Value

The command is used for its side-effect of generating a plot

Author(s)

N. Ellis, CSIRO, Cleveland, Australia. <[e-mail]>

References

Ellis, N., Smith, S.J., and Pitcher, C.R. (2012). Gradient Forests: calculating importance gradients on physical predictors. Ecology, 93, 156–168.

See Also

predict.gradientForest, gradientForest

Examples

data(GBRexample)

# transform the predictors using predict() on a fine-scale grid
predictors <- names(importance(gf.GBR))
gf.pred <- predict(gf.GBR, Phys_grid[,predictors])
plot(gf.pred, Phys_grid[,c("EAST")], Phys_grid[,c("NORTH")], asp=1, palette="gr", pch=15, 
main="Biological composition in Far North Great Barrier Reef")

gradientForest documentation built on Aug. 24, 2023, 3:03 p.m.