GA: Genetic Algorithms package for STAT243 course

plot_final <- function(input){
  #' Plot goodness of fit vs generation
  #'
  #' This function works to plot goodness of fit against generation
  #'
  #' @author Xiao Li
  #'
  #' @param input It is a list. The first element of the list contains all values of goodness of fit
  #' during different generations.
  #'
  require(ggplot2)
  m <- matrix(unlist(input[[1]]),nrow = length(input[[1]]),byrow = T)
  pts <- data.frame()
  for(i in 1:nrow(m)){
    pts <- rbind(pts,data.frame(x=rep(i,length(m[i,])),y=m[i,]))
  }

  g <- ggplot(data = pts, aes(x=x,y=y)) +
    geom_point() +
    xlab("Generations") +
    ylab("Goodness of Fit") +
    stat_summary(aes(y = y,group=1), fun.y=mean, colour="red", geom="line",group=1)
  return(g)
}

# # Fixing all parameters except mutation rate
# mutation_001 <- select(data=mtcars,
#                      method = "onepropselection",
#                      p=2,
#                      mutation_rate = 0.01,
#                      regression_target = "mpg",
#                      scheme = "re-rank",
#                      max_iter = 500)
#
# mutation_010 <- select(data=mtcars,
#                      method = "onepropselection",
#                      p=2,
#                      mutation_rate = 0.1,
#                      regression_target = "mpg",
#                      scheme = "re-rank",
#                      max_iter = 500)
#
# plot_final(mutation_001)
# plot_final(mutation_010)
#
# # Fixing all parameters except number of partitions in crossover
# p_2 <- select(data=mtcars,
#             method = "onepropselection",
#             p=2,
#             mutation_rate = 0.01,
#             regression_target = "mpg",
#             scheme = "re-rank",
#             max_iter = 500)
#
# p_5 <- select(data=mtcars,
#             method = "onepropselection",
#             p=5,
#             mutation_rate = 0.01,
#             regression_target = "mpg",
#             scheme = "re-rank",
#             max_iter = 500)
#
# plot_final(p_2)
# plot_final(p_5)
#
# # Fixing all parameters except method of selecting parents
# select_1 <- select(data=mtcars,
#                  method = "onepropselection",
#                  p=2,
#                  mutation_rate = 0.01,
#                  regression_target = "mpg",
#                  scheme = "re-rank",
#                  max_iter = 500)
#
# select_2 <- select(data=mtcars,
#                  method = "twopropselection",
#                  p=2,
#                  mutation_rate = 0.01,
#                  regression_target = "mpg",
#                  scheme = "re-rank",
#                  max_iter = 500)
#
# select_3 <- select(data=mtcars,
#                  method = "tournament",
#                  p=2,
#                  mutation_rate = 0.01,
#                  regression_target = "mpg",
#                  scheme = "re-rank",
#                  max_iter = 500)
#
# plot_final(select_1)
# plot_final(select_2)
# plot_final(select_3)
#
# # Fixing all parameters except method of replacing population
# replace_prop <- select(data=mtcars,
#                      method = "tournament",
#                      p=2,
#                      mutation_rate = 0.01,
#                      regression_target = "mpg",
#                      scheme = "proportion",
#                      max_iter = 500)
#
# replace_rerank <- select(data=mtcars,
#                        method = "tournament",
#                        p=2,
#                        mutation_rate = 0.01,
#                        regression_target = "mpg",
#                        scheme = "re-rank",
#                        max_iter = 500)
#
# plot_final(replace_prop)
# plot_final(replace_rerank)