ELEFAN_GA_boot: Bootstrapped ELEFAN_GA

In rschwamborn/fishboot: Bootstrap-based methods for the study of fish stocks and aquatic populations

Description

'ELEFAN_GA_boot' performs a bootstrapped fitting of von Bertalanffy growth function (VBGF) via the ELEFAN_GA function. Most of the arguments are simply passed to the function within many permutations (resampling) of the original lfq data.

Usage

ELEFAN_GA_boot(lfq, seasonalised = FALSE, low_par = NULL,
  up_par = NULL, parallel = TRUE, nresamp = 200,
  no_cores = detectCores() - 1, clusterType = "PSOCK",
  outfile = "output.txt", popSize = 60, maxiter = 50, run = 200,
  pmutation = 0.2, pcrossover = 0.8, elitism = base::max(1,
  round(popSize * 0.05)), MA = 5, addl.sqrt = FALSE, agemax = NULL,
  flagging.out = TRUE, resample = TRUE, seed = NULL)

Arguments

lfq	a length frequency object of the class 'lfq'
seasonalised	logical; indicating if the seasonalised von Bertalanffy growth function should be applied (default: FALSE).
low_par	a list providing the minimum of the search space in case of real-valued or permutation encoded optimizations. When set to NULL the following default values are used: Linf length infinity in cm (default is calculated from maximum length class in the data), K curving coefficient (default: 0.01), t_anchor time point anchoring growth curves in year-length coordinate system, corrsponds to peak spawning month (range: 0 to 1, default: 0), C amplitude of growth oscillation (range: 0 to 1, default: 0), ts summer point (ts = WP - 0.5) (range: 0 to 1, default: 0);
up_par	a list providing the maximum of the search space in case of real-valued or permutation encoded optimizations. When set to NULL the following default values are used: Linf length infinity in cm (default is calculated from maximum length class in the data), K curving coefficient (default: 1), t_anchor time point anchoring growth curves in year-length coordinate system, corrsponds to peak spawning month (range: 0 to 1, default: 1), C amplitude of growth oscillation (range: 0 to 1, default: 1), ts summer point (ts = WP - 0.5) (range: 0 to 1, default: 1);
parallel	logical; should parallelized computing be used. This differs from the 'parallel' argument in ELEFAN_GA in that it is not used within the ga function for calculation at the population level, but rather for each resampling. Depending on platform operating system, the argument 'clusterType' can be adjusted (see argument description for details). (Default: 'parallel = TRUE')
nresamp	numeric; the number of permutations to run (Default: 'nresamp = 200')
no_cores	numeric (Default: 'no_cores = detectCores() - 1')
clusterType	(Default: 'clusterType = "PSOCK"')
outfile	character; text file name (Default: 'outfile = "output.txt"') which will records the progress of the permutation completions.
popSize	the population size. (Default: 'popSize = 60')
maxiter	the maximum number of iterations to run before the GA search is halted. (Default: 'maxiter = 50')
run	the number of consecutive generations without any improvement in the best fitness value before the GA is stopped. (Default: 'run = 10')
pmutation	numeric. A small fraction of 1.0. The probability of mutation in a parent chromosome. Usually mutation occurs with a small probability. (Default: 'pmutation = 0.2')
pcrossover	the probability of crossover between pairs of chromosomes. Typically this is a large value. (Default: 'pcrossover = 0.8')
elitism	the number of best fitness individuals to survive at each generation. By default the top 5% individuals will survive at each iteration.
MA	number indicating over how many length classes the moving average should be performed (default: 5, for more information see lfqRestructure
addl.sqrt	logical. Should counts be square root transformed prior to restructuring.
agemax	numeric. maximum age
flagging.out	logical Should flagging out be done
resample	logical. Should 'lfq' object be resampled (Default: 'resample = TRUE'). When 'resample = FALSE', a 'partial bootstrap' is performed, reflecting solution variation due only to the search algorithm.
seed	seed value for random number reproducibility (Default: NULL)

Value

a list of class 'lfqBoot' containing 2 levels: '$bootRaw' - a data.frame of fitted VBGF parameters (columns) by resampling (rows), '$seed' - a vector of seed values set prior to each resampling call to 'lfqResample'.

Examples

# load data
library(TropFishR)
data(alba)

# settings (these settings may not be optimal - for demo only)
MA <- 7
low_par <- list(Linf = 8, K = 0.1, t_anchor = 0, C = 0, ts = 0)
up_par <- list(Linf = 15, K = 5, t_anchor = 1, C = 1, ts = 1)
popSize <- 60
maxiter <- 50
run <- 10
pmutation <- 0.2
nresamp <- 12


# parallel version
library(parallel)
t1 <- Sys.time()
res <- ELEFAN_GA_boot(lfq=alba, MA = MA, seasonalised = FALSE,
  up_par = up_par, low_par = low_par,
  popSize = popSize, maxiter = maxiter, run = run, pmutation = pmutation,
  nresamp = nresamp, parallel = TRUE, no_cores = detectCores()-1,
  seed = 1
)
t2 <- Sys.time()
t2 - t1
res


# non-parallel version
t1 <- Sys.time()
res <- ELEFAN_GA_boot(lfq=alba, seasonalised = FALSE,
  up_par = up_par, low_par = low_par,
  popSize = popSize, maxiter = maxiter, run = run,
  pmutation = pmutation, nresamp = nresamp, MA = MA, parallel = FALSE,
  seed = 1
)
t2 <- Sys.time()
t2 - t1
res

# plot resulting distributions
univariate_density(res, use_hist = TRUE)

# plot scatterhist of Linf and K
LinfK_scatterhist(res)

rschwamborn/fishboot documentation built on July 27, 2019, 12:36 a.m.