Filled.Contour in R 3.0.X Throws Error

filled.contour in R 3.0.x throws error

This happens if you use a non-standard API. You are allowed to do that, but cannot expect that it is maintained.

Change

.Internal(filledcontour(as.double(x), as.double(y), z, as.double(levels), 
                                                        col = col))

to

.filled.contour(as.double(x), as.double(y), z, as.double(levels), 
                            col = col)

The change was announced with the release notes:

The C code underlying base graphics has been migrated to the graphics
package (and hence no longer uses .Internal() calls).

Have you ever heard of a "minimal reproducible example" (emphasis on "minimal")?

Placing rows at user-defined distances in filled.contour in R

While I waited for some response here, I kept on trying and actually it was a very silly problem.
You just need to change the default y argument in filled.contour

Here is the (not significantly) modified function which gives the following image:

filled.contour(z=z1,y=c(0,.33,1),col=col1,cex.lab=2,cex.main=1.1,nlevels=20,main=('Heat map with custom y placement'))

Multiple filled.contour plots in one graph using with par(mfrow=c())

No I do not think this is possible in filled.contour.

Although extensions have been written for you already. To be found here, here and here and a legend code here.
[If you are using the filled.contour3 function referred to on those sites, and using a more recent version, then you need to use the upgrade fix referred to in this SO post].
Using those codes I produced:

Creating a colour gradient around zero in filled.contour

I have only managed to achieve what you want using ggplot2.

You can try the following (I submerged the volcano data as an example):

library(ggplot2)
library(reshape2)

## Just an example, I subtract the mean to have positive and negative values for z 
dd <- volcano-mean(volcano)

## Creates a data.frame with columns x, y, z
dd <- melt(dd)
names(dd) <- c('x','y','z')

## Does the contour plot
d <- ggplot(dd, aes(x,y,z=z))
d + geom_tile(aes(fill=z))  + scale_fill_gradient2(low="blue", high="red")

---

I wrote a small function that does what you want to achieve:

myFilled.contour <- function(x = seq(0, 1, length.out = nrow(z)),
                             y = seq(0, 1, length.out = ncol(z)),
                             z, nlevels=30, ...) {
  ma <- max(abs(z))
  lvls <- seq(-ma, ma, length.out = nlevels)
  cols <- colorRampPalette(c("blue","white","red")) (nlevels - 1)
  filled.contour(x, y, z, plot.axes={axis(1); axis(2)},
                 col=cols, levels=lvls, ...)
}

Using filled.contour and again the submerged volcano dd:

myFilled.contour(z=d)

Using your data:

myFilled.contour(x,y,z)

Caveat: The legend includes levels not used in the contour plot.

Hope it helps,

alex

Error when plotting RasterLayer in R 3.x.x

The problem could be solved by starting a new R-Studio project in a new directory!

First I thought I understand, because it is always possible, that some of my own functions in the environment could have be kind of conflicting, since they were created before 3.x.x.

But then I realized, that I still do not understand , because clearing the environment and history did not help in the first place.

Since I am a curious person, I'm still interested in ideas, which can explain that and I have absolutely no understanding for voting down my question without giving an explanation!

Filled contour plot with R/ggplot/ggmap

It's impossible to test without a more representative dataset (can you provide a link?).

Nevertheless, try:

## not tested..
map = ggmap( baseMap ) + 
    stat_contour( data = flood, geom="polygon", 
                  aes( x = lon, y = lat, z = rain, fill = ..level.. ) ) +
    scale_fill_continuous( name = "Rainfall (inches)", low = "yellow", high = "red" ) 

The problem is that geom_contour doesn't respect fill=.... You need to use stat_contour(...) with geom="polygon" (rather than "line").

increasing 'x' and 'y' values expected

As the error message implies, one problem is that in y<-c(10:1) the y values do not increase...

Besides, z needs to be an x*y matrix.

So try the following:

x<-c(1:10)
y<-c(1:10)
z<-matrix(x,nrow=length(x),ncol=length(y)

persp(x,y,z, theta = 30, phi = 30, expand = 0.5, col = "lightblue")

---

Related Topics