Introduction:

This document will try some plotting of data across full sets of runs.

Setup code

library('ggplot2')
library('cluster')
library('Hmisc')
library("gridExtra")
#library('apcluster')

setwd("~/Documents/R/Clustering/lexicase-clusturing-analysis/R_notebooks")

source('../scripts/clustering.R')

Replace Space With Newline + Lexicase

Import the Data

# Copied from SO
file_list = list.files(path="../data/RSWN/lexicase/clustering/", pattern="*.csv")
myfiles = lapply(paste("../data/RSWN/lexicase/clustering/", file_list, sep=""), read.csv)

all_data = do.call(rbind, myfiles)

Some plots!

plot(all_data$generation, all_data$cluster.count, pch=3, cex=0.3, col=rgb(0.3,0.4,0.6,0.5), ylim=c(0, 350))

plot(all_data$generation, all_data$error.diversity, pch=3, cex=0.3, col=rgb(0.3,0.4,0.6,0.5), ylim=c(0, 1))

Replace Space With Newline + Tourney size 7

Import the Data

# Copied from SO
file_list_tourney = list.files(path="../data/RSWN/tourney/clustering/", pattern="*.csv")
myfiles_tourney = lapply(paste("../data/RSWN/tourney/clustering/", file_list_tourney, sep=""), read.csv)

all_data_tourney = do.call(rbind, myfiles_tourney)

Some plots!

plot(all_data_tourney$generation, all_data_tourney$cluster.count, pch=3, cex=0.3, col=rgb(0.3,0.4,0.6,0.5), ylim=c(0, 350))

plot(all_data_tourney$generation, all_data_tourney$error.diversity, pch=3, cex=0.3, col=rgb(0.3,0.4,0.6,0.5), ylim=c(0, 1))

Replace Space With Newline + IFS size 7

Import the Data

# Copied from SO
file_list_ifs = list.files(path="../data/RSWN/ifs/clustering/", pattern="*.csv")
myfiles_ifs = lapply(paste("../data/RSWN/ifs/clustering/", file_list_ifs, sep=""), read.csv)

all_data_ifs = do.call(rbind, myfiles_ifs)

Some plots!

plot(all_data_ifs$generation, all_data_ifs$cluster.count, pch=3, cex=0.3, col=rgb(0.3,0.4,0.6,0.5), ylim=c(0, 350))

plot(all_data_ifs$generation, all_data_ifs$error.diversity, pch=3, cex=0.3, col=rgb(0.3,0.4,0.6,0.5), ylim=c(0, 1))

Plot medians

all_rswn_data = rbind(all_data, all_data_tourney, all_data_ifs)

all_rswn_data$treatment = factor(all_rswn_data$treatment, levels = c("lexicase", "tourney", "ifs"))

rswn_div_plot <- ggplot(all_rswn_data, aes(x=generation, y=error.diversity,
                                                            color=treatment)) + 
  stat_summary(fun.data="median_hilow", conf.int=0.5, alpha=0.5) + theme(axis.title.x=element_blank())
  #facet_grid(treatment ~ .)

count_successes_for_gen <- function(data, gen) {
  return(100-nrow(subset(data, generation==gen)))
  }

count_all_successes <- function(data) {
  return(sapply(seq(0, 300), function (gen) count_successes_for_gen(data, gen)))
}

lexicase_success_counts = data.frame(generation=seq(0, 300), num.successes=count_all_successes(all_data))
lexicase_success_counts$treatment = 'lexicase'
tourney_success_counts = data.frame(generation=seq(0, 300), num.successes=count_all_successes(all_data_tourney))
tourney_success_counts$treatment = 'tourney'
ifs_success_counts = data.frame(generation=seq(0, 300), num.successes=count_all_successes(all_data_ifs))
ifs_success_counts$treatment = 'ifs'

success_counts = rbind(lexicase_success_counts, tourney_success_counts, ifs_success_counts)
success_counts$treatment = factor(success_counts$treatment, levels = c("lexicase", "tourney", "ifs"))

rswn_lex_success_plot <- ggplot(success_counts, aes(x=generation, y=num.successes, color=treatment)) + geom_line() + ylim(c(0,100))

grid.arrange(arrangeGrob(rswn_div_plot, rswn_lex_success_plot, heights=c(3/4, 1/4), ncol=1))

ggplot(rbind(all_data, all_data_tourney, all_data_ifs), aes(x=generation, y=cluster.count,
                                                            color=interaction(succeeded, treatment))) + 
  stat_summary(fun.data="median_hilow", conf.int=0.5, alpha=0.8)

  #facet_grid(treatment ~ .)

Syllables + Lexicase

Import the Data

# Copied from SO
file_list_syllables_lexicase = list.files(path="../data/syllables/lexicase/clustering/", pattern="*.csv")
files_syllables_lexicase = lapply(paste("../data/syllables/lexicase/clustering/", file_list_syllables_lexicase, sep=""), read.csv)

syllables_lexicase = do.call(rbind, files_syllables_lexicase)

Some plots!

plot(syllables_lexicase$generation, syllables_lexicase$cluster.count, pch=3, cex=0.3, col=rgb(0.3,0.4,0.6,0.5),
     ylim=c(0, 700))

plot(syllables_lexicase$generation, syllables_lexicase$error.diversity, pch=3, cex=0.3, col=rgb(0.3,0.4,0.6,0.5), ylim=c(0, 1))

Syllables + IFS Size 7

Import the Data

# Copied from SO
file_list_syllables_ifs = list.files(path="../data/syllables/ifs/clustering/", pattern="*.csv")
files_syllables_ifs = lapply(paste("../data/syllables/ifs/clustering/", file_list_syllables_ifs, sep=""), read.csv)

syllables_ifs = do.call(rbind, files_syllables_ifs)

Some plots!

plot(syllables_ifs$generation, syllables_ifs$cluster.count, pch=3, cex=0.3, col=rgb(0.3,0.4,0.6,0.5),
     ylim=c(0, 700))

plot(syllables_ifs$generation, syllables_ifs$error.diversity, pch=3, cex=0.3, col=rgb(0.3,0.4,0.6,0.5), ylim=c(0, 1))

To Do