r - ggplot2: can a non-aesthetic parameter vary by factor? -

i want create ggplot in statistical parameter varies according aesthetically mapped factor. specifically, i'd create contour plot using stat_density2d(), i'd to map discrete factor color, , i'd specify different break values each factor level.

here's minimal working example:

d <- data.frame(x=c(rnorm(500), rnorm(500, 2)), y=rnorm(1000), z=c(rep("a", 500), rep("b", 500)))  ggplot(d, aes(x, y, fill=z)) +        stat_density2d(breaks=.05, geom="polygon", alpha=.5) 

this i'm going except breaks identical across factors. i'm looking way specify different break values each factor. 1 way create separate layer each factor:

ggplot() +        stat_density2d(data=d[d$z=="a",], aes(x, y), breaks=.05, geom="polygon", alpha=.5, fill="red") +       stat_density2d(data=d[d$z=="b",], aes(x, y), breaks=.1, geom="polygon", alpha=.5, fill="blue") 

but isn't workable because lose legend , it's cumbersome cases more 2 factor levels.

i'd combine strengths of above 2 cases. i'm positive i've seen way accomplish i'm trying do, can't seem find it's relevant. have thoughts on possible solution?

remebered couple of years ago - taking solution directly kohske's answer

# data set.seed(1)     d <- data.frame(x=c(rnorm(500), rnorm(500, 2)), y=rnorm(1000),                                     z=c(rep("a", 500), rep("b", 500)))  library(plyr) library(ggplot2)  hls <- mapply(function(x, b) stat_density2d(data = x, breaks = b,                      geom="polygon", alpha=.5), dlply(d, .(z)), c(0.05, 0.1))  ggplot(d, aes(x, y, fill=z)) + hls + scale_fill_manual(values=c("red", "blue")) 

# compare ggplot() +    stat_density2d(data=d[d$z=="a",], aes(x, y), breaks=.05, geom="polygon",                                                        alpha=.5, fill="red") +   stat_density2d(data=d[d$z=="b",], aes(x, y), breaks=.1, geom="polygon",                                                      alpha=.5, fill="blue")