as.camodel_initmat {chouca}

Convert a matrix to a CA model landscape

Description

Convert a matrix to a CA model landscape for later use with run_camodel or run_meanfield.

Usage

as.camodel_initmat(m, levels = NULL)

Arguments

m

The input matrix (numeric, character or factor)

levels

The levels to use in the resulting landscape. If NULL, the unique values of the input matrix are used as levels. Set this manually if you want the resulting landscape to have extra levels that are not present in the original matrix.

Value

This function returns a matrix containing values as factors, with levels corresponding to the levels argument. This matrix has the class camodel_initmat so that it can be displayed with the image generic function and works well with CA-related functions (such as run_camodel).

See Also

generate_initmat, run_camodel, run_meanfield

Examples

# Simple conversion of a matrix with regular patterns
x <- seq(0, 2 * pi, l = 256)
z <- outer(x, x, function(x, y) as.numeric(sin(10*x) + cos(10*y) > 0.8))
mat <- as.camodel_initmat(z)
summary(mat)
image(mat)

# This is a character matrix. We need to convert it to use it as input to
# run_camodel()
size <- 64
m <- matrix(ifelse(runif(size^2) < .5, "TREE", "EMPTY"), nrow = size, ncol = size)
m <- as.camodel_initmat(m)
summary(m) # this is a landscape object
image(m)

# Start a simulation using this matrix
mod <- ca_library("forestgap")
out <- run_camodel(mod, m, seq(0, 256))
plot(out)


# Run a glider in the game of life
mod <- ca_library("gameoflife")
init <- matrix(c(0, 0, 1, 0, 0, 0, 0,
                 0, 0, 0, 1, 0, 0, 0,
                 0, 1, 1, 1, 0, 0, 0,
                 0, 0, 0, 0, 0, 0, 0,
                 0, 0, 0, 0, 0, 0, 0,
                 0, 0, 0, 0, 0, 0, 0),
                nrow = 6, ncol = 7, byrow = TRUE)
init[] <- ifelse(init == 1, "LIVE", "DEAD")
# image() does not work on init here without conversion by as.camodel_initmat
init <- as.camodel_initmat(init)
image(init)

# Run the model and display simulation output as it is running
ctrl <- list(custom_output_fun = landscape_plotter(mod, fps_cap = 5),
             custom_output_every = 1)
out <- run_camodel(mod, init, times = seq(0, 32), control = ctrl)