R/evolve.R

Defines functions evolve

Documented in evolve

#' Move the simulation forward one or more steps
#'
#' This is the function that move the simulation forward in time. It is possible
#' to either specify the number of steps that should be simulated or let the
#' simulation terminate as `alpha_min` is reached. Note that some values of
#' `alpha` and `alpha_target` does not allow alpha to converge to `alpha_min` so
#' letting the simulation self-terminate can result in an infinite loop. The
#' default settings will result in `alpha_min` being reached in 300 generations.
#'
#' @details
#' Each generation in the simulation progress in the following manner:
#'
#' 1. Check whether the specified number of generations has been reached
#' 2. Check whether `alpha_min` has been reached
#' 3. If either 1. or 2. is true, terminate the simulation
#' 4. Apply the forces on the current particle positions and velocities in the
#' order they have been added
#' 5. Reduce the velocity according to the given `velocity_decay`
#' 6. Update the position and velocity based on any provided constraints
#' 7. Calculate the new particle positions based on the new velocity
#' 8. If given, call the `on_generation` function.
#'
#' @param simulation A simulation object
#' @param steps The number of generations to progress or a function getting the
#' simulation object and returns `TRUE` if the simulation should proceed and
#' `FALSE` if it should stop. If `NULL` the simulation will run until
#' `alpha_min` has been reached.
#' @param on_generation A function to be called after each generation has been
#' progressed. The function will get the current state of the simulation as the
#' first argument. If the function returns a simulation object it will replace
#' the current simulation from the next generation. In the case of any other
#' return type the return will be discarded and the function will have no effect
#' outside its side-effects.
#' @param ... Additional arguments to `on_generation`
#'
#' @return A simulation object with updated positions and velocities
#'
#' @export
#'
#' @examples
#' graph <- tidygraph::create_notable('folkman')
#' sim <- graph %>%
#'   simulate() %>%
#'   wield(link_force) %>%
#'   wield(manybody_force)
#'
#' # Take 5 steps and tell about it
#' sim %>% evolve(5, function(sim) {
#'   cat('Generation: ', evolutions(sim), '\n', sep = '')
#' })
#'
#' # Run evolution until alpha_min is reached
#' sim %>% evolve(NULL)
#'
evolve <- function(simulation, steps = NULL, on_generation = NULL, ...) {
  stopifnot(is.simulation(simulation))
  while (TRUE) {
    if (!is.null(steps)) {
      if (is.function(steps)) {
        if (!steps(simulation)) break
      } else {
        steps <- steps - 1
        if (steps < 0) break
      }
    }
    if (alpha(simulation) < universe_def(simulation)$alpha_min) break
    particles <- particles(simulation)
    alpha <- alpha(simulation, progress = TRUE)
    alpha(simulation) <- alpha

    evolution <- Reduce(function(l ,r) {
      res <- apply_force(r, particles, l$position, l$velocity, alpha)
      if (all(r$include)) return(res)
      l$position[r$include, ] <- res$position[r$include, , drop = FALSE]
      l$velocity[r$include, ] <- res$velocity[r$include, , drop = FALSE]
      l
    }, forces(simulation), init = list(position = position(simulation), velocity = velocity(simulation)))
    velocity(simulation) <- evolution$velocity * (1 - universe_def(simulation)$velocity_decay)

    constrained_evolution <- Reduce(function(l ,r) {
      res <- apply_constraint(r, particles, l$position, l$velocity, alpha)
      if (all(r$include)) return(res)
      l$position[r$include, ] <- res$position[r$include, , drop = FALSE]
      l$velocity[r$include, ] <- res$velocity[r$include, , drop = FALSE]
      l
    }, constraints(simulation), init = list(position = position(simulation), velocity = velocity(simulation)))
    velocity(simulation) <- constrained_evolution$velocity
    position(simulation) <- constrained_evolution$position

    position(simulation) <- position(simulation) + velocity(simulation)

    simulation <- advance(simulation)
    if (!is.null(on_generation)) {
      tmp_sim <- on_generation(
        simulation,
        ...
      )
      if (is.simulation(tmp_sim)) simulation <- tmp_sim
    }
  }
  simulation
}

Try the particles package in your browser

Any scripts or data that you put into this service are public.

particles documentation built on May 2, 2019, 4:01 p.m.