generate_initmat {chouca}

Generate an initial matrix for a chouca model

Description

Helper function to create a spatially-random initial landscape (matrix) with specified covers for a cellular automaton

Usage

generate_initmat(mod, pvec, nrow, ncol = nrow)

Arguments

mod

A stochastic cellular automaton model created by camodel

pvec

A numeric vector of covers for each state in the initial configuration, possibly with named elements.

nrow

The number of rows of the output matrix

ncol

The number of columns of the output matrix

Details

This function is a helper to build a starting configuration (matrix) for a stochastic cellular automaton based on the definition of the model and the specified starting covers (in pvec). It will produce a landscape with expected global cover of each state equal to the covers in pvec, and a completely random spatial structure.

The length of the pvec vector must match the number of possible cell states in the model. If present, the names of pvec must match the states defined in the model. In this case, they will be used to determine which state gets which starting cover instead of the order of the values.

The pvec will be normalized to sum to one, and the function will produce a warning if this produces a meaningful change in covers.

If you already have a matrix you want to use as a starting configuration, we recommend you to use as.camodel_initmat to convert it to an object that run_camodel can use.

Value

This function returns a matrix containing values as factors, with levels corresponding to the model states (defined in the mod argument) and dimensions set by nrow and ncol. This matrix has the class camodel_initmat so that it can be displayed with the image generic function.

See Also

as.camodel_initmat

Examples

# Run the Game of Life starting from a random grid
game_of_life <- ca_library("gameoflife")
grid <- generate_initmat(game_of_life, c(LIVE = .1, DEAD = .9), nrow = 64)
out <- run_camodel(game_of_life, grid, times = seq(0, 128))
image(out) # final configuration

# Logistic growth of plants
mod <- camodel(
  transition(from = "empty", to = "plant", ~ r * p["plant"]),
  transition(from = "plant", to = "empty", ~ m),
  parms = list(r = 1, m = .03),
  wrap = TRUE,
  neighbors = 8
)
grid <- generate_initmat(mod, c(empty = .99, plant = .01), nrow = 128)
image(grid) # initial state
out <- run_camodel(mod, grid, times = seq(0, 30))
image(out) # final state
plot(out) #