| 1 | # find nearest neighbors for GP indiviuals
|
|---|
| 2 |
|
|---|
| 3 | library(ggplot2);
|
|---|
| 4 |
|
|---|
| 5 | gp_log <- read.csv2("C:/Users/P24581/filebox/GPTP 2018/symbreg-models-13.05.2018-1418.csv",header = FALSE,sep='\t',dec=',');
|
|---|
| 6 | idx <- seq(1:nrow(gp_log));
|
|---|
| 7 | # check popSize in gp_log
|
|---|
| 8 | gp_log[seq(1,34000,500),1]
|
|---|
| 9 | popSize <- 500;
|
|---|
| 10 |
|
|---|
| 11 | # check qualities
|
|---|
| 12 | # ggplot(gp_log, aes(x=idx,y=gp_log$V1)) + geom_line();
|
|---|
| 13 |
|
|---|
| 14 | # generations <- seq(1,34000/popSize,1);
|
|---|
| 15 | generations <- seq(1,15,1);
|
|---|
| 16 | numClusters <- max(m$c);
|
|---|
| 17 |
|
|---|
| 18 | gp_evals <- gp_log[,seq(3,202,2)];
|
|---|
| 19 | all_evals <- m[,6:105];
|
|---|
| 20 | for(gen_i in generations) {
|
|---|
| 21 | #gen_i <- 15;
|
|---|
| 22 | selectedRows <- seq((gen_i - 1)*popSize + 1, gen_i * popSize,1);
|
|---|
| 23 | min(selectedRows)
|
|---|
| 24 | max(selectedRows)
|
|---|
| 25 | xcorrel <- cor(t(all_evals[,]), t(gp_evals[selectedRows,]))^2
|
|---|
| 26 | mapped_gp_log <- m[max.col(t(xcorrel)), 1:5]
|
|---|
| 27 | #check
|
|---|
| 28 | #cor(t(all_evals[128082,]), t(gp_evals[2,]))^2
|
|---|
| 29 | #max(cor(t(all_evals[,]), t(gp_evals[2,]))^2)
|
|---|
| 30 | ggplot(mapped_gp_log, aes(x=x, y=y)) + xlim(-75,75) + ylim(-75,75) + geom_point();
|
|---|
| 31 | ggsave(paste("scatter",gen_i,".png"))
|
|---|
| 32 |
|
|---|
| 33 | ggplot(mapped_gp_log, aes(x=c)) +xlim(0,numClusters+1) + geom_histogram(binwidth = 1);
|
|---|
| 34 | ggsave(paste("cluster_freq",gen_i,".png"))
|
|---|
| 35 | }
|
|---|
