Rewrite that R code to Julia: #"C:\Users\Michał\Desktop\player22.csv"…
- User
Rewrite that R code to Julia: #"C:\Users\Michał\Desktop\player22.csv" Data<-read.csv("C:/Users/Michał/Desktop/player22.csv")
RW_idx <- which(grepl("RW", DataPositions)) GK_idx <- which(grepl("GK", DataPositions)) LW_idx <- which(grepl("LW", DataPositions)) LM_idx <- which(grepl("LM", DataPositions)) CB_idx <- which(grepl("CB", DataPositions)) LB_idx <- which(grepl("LB", DataPositions)) RM_idx <- which(grepl("RM", DataPositions)) RWB_idx <- which(grepl("RWB", Data$Positions))
#############
position_vectors <- list(RW_idx, ST_idx, GK_idx, CM_idx, LW_idx, CDM_idx, LM_idx, CF_idx, CB_idx, CAM_idx, LB_idx, RB_idx, RM_idx, LWB_idx, RWB_idx)
position_vectors_list<-position_vectors
##############
Mutation 2
mutate <- function(selected_players_df, position_vectors_list, probability) { n_rows <- nrow(selected_players_df) selected_players_matrix <- selected_players_df
select_random_player <- function(idx_list, selected_players = NULL) { repeat { random_idx <- sample(idx_list, 1) if (!random_idx %in% selected_players) { return(random_idx) } } }
for (i in 1:n_rows) { for (pos_idx in 1:length(position_vectors_list)) { if (runif(1) <= probability) { selected_players_matrix[i, pos_idx] <- select_random_player(position_vectors_list[[pos_idx]], selected_players = selected_players_matrix[i, ]) } } }
selected_players_df <- data.frame(selected_players_matrix) return(selected_players_df) }
n_rows<-100
pop_init<- as.data.frame(matrix(NA, n_rows, length(position_vectors))) pop_init<- mutate(pop_init, position_vectors_list, probability=1)
#######Target
row1<-pop_init[1,]
target <- function(row1,penalty=5) { position_ratings <- c("RWRating", "STRating", "GKRating", "CMRating", "LWRating", "CDMRating", "LMRating", "CFRating", "CBRating", "CAMRating", "LBRating", "RBRating", "RMRating", "LWBRating", "RWBRating")
row1<-as.matrix(row1,nrow=1) parent_data <- Data[row1, ] ratings <- parent_data[position_ratings] ratings_log <- log(ratings)
potential_minus_age <- 0.15*parent_dataAge int_reputation <- parent_data$IntReputation
sumratings<-0 rating_list<-c() for (i in 1:15){ temp<-ratings_log[i,i] sumratings<-sumratings+temp rating_list<-append(rating_list, temp)
}
Apply constraints
constraint_penalty <- 0 if (sum(parent_dataValueEUR) > 250000000) { constraint_penalty <- constraint_penalty + log((sum(parent_dataValueEUR)-250000000)^penalty) } if (sum(parent_dataWageEUR) > 250000) { constraint_penalty <- constraint_penalty + log((sum(parent_dataWageEUR)-250000)^penalty) } if (any(rating_list < 1.2)) { constraint_penalty <- constraint_penalty + 1.2^penalty }
potential_minus_age
target_value <- -(sumratings+0.3*sum(potential_minus_age) +sum(int_reputation))+constraint_penalty
return(target_value) }
#target(global_best)
parents<-pop_init t_size=2 penalty=1
tournament_selection <- function(parents, t_size,penalty=6) { random_parents_idx <- sample(nrow(parents), t_size, replace = FALSE) random_parents <- parents[random_parents_idx, ]
random_parents_fitness <- apply(random_parents, 1, function(x) target(x, penalty=penalty))
best_parent_idx <- which.min(random_parents_fitness) return(random_parents[best_parent_idx, ]) }
tournament_selection (pop_init,t_size=2)
#best parent 1 #tournement_selection *50
corssover_point<-6
parent1<-row1 parent2<-pop_init[2,]
crossover <- function(parent1, parent2,corssover_point) {
offspring1 <- c(parent1[1:crossover_point], parent2[(crossover_point + 1):ncol(parent1)]) offspring2 <- c(parent2[1:crossover_point], parent1[(crossover_point + 1):ncol(parent2)])
return(rbind(offspring1, offspring2)) }
crossover_point<-7 population_size <- 100 num_generations <- 20000 tournament_size <- 2 probability<-0.09 penalty<-1 parents<-pop_init
global_best<-pop_init[1,] global_best_value<-target(global_best)
for (gen in 1:num_generations) {
parent_pop<- as.data.frame(matrix(NA, n_rows, length(position_vectors))) for(c in 1:(population_size)){ parent_pop[c,]<-tournament_selection (parents,t_size=2,penalty=penalty)
}
offspring_temp<- as.data.frame(matrix(NA, 1, length(position_vectors)))
for (c in seq(1, population_size, by = 2)){ parent1<-parent_pop[c,] parent2<-parent_pop[c+1,]
offsprings<-crossover(parent1, parent2,corssover_point) o1<-unlist(offsprings[1,]) o2<-unlist(offsprings[2,]) offspring_temp<-rbind(offspring_temp,o1) offspring_temp<-rbind(offspring_temp,o2)} offspring_temp <- offspring_temp[-1, ]
parents<- mutate(offspring_temp, position_vectors_list, probability=probability) row.names(parents) <- 1:population_size solutions<-apply(parents, 1, target) idx_sol<-which.min(solutions)[[1]] temp_best<-parents[idx_sol[[1]],] temp_target_value<-solutions[idx_sol[[1]]]
if(temp_target_value<=global_best_value){ global_best<-temp_best global_best_value<-temp_target_value
}
penalty<-penalty+0.5 if(penalty>=4){ penalty<-0 }
}
global_best_fin<-as.matrix(global_best,nrow=1) parent_data <- Data[global_best_fin, ]
target(global_best)
Be as careful as possible.
- Assistant
Load packages
using CSV, DataFrames, Random, StatsBase, LinearAlgebra
Read data
Data = CSV.read("C:/Users/Michał/Desktop/player22.csv", DataFrame)
Indices of players for each position
RW_idx = findall(x -> occursin("RW", x), Data[!, :Positions]) ST_idx = findall(x -> occursin("ST", x), Data[!, :Positions]) GK_idx = findall(x -> occursin("GK", x), Data[!, :Positions]) CM_idx = findall(x -> occursin("CM", x), Data[!, :Positions]) LW_idx = findall(x -> occursin("LW", x), Data[!, :Positions]) CDM_idx = findall(x -> occursin("CDM", x), Data[!, :Positions]) LM_idx = findall(x -> occursin("LM", x), Data[!, :Positions]) CF_idx = findall(x -> occursin("CF", x), Data[!, :Positions]) CB_idx = findall(x -> occursin("CB", x), Data[!, :Positions]) CAM_idx = findall(x -> occursin("CAM", x), Data[!, :Positions]) LB_idx = findall(x -> occursin("LB", x), Data[!, :Positions]) RB_idx = findall(x -> occursin("RB", x), Data[!, :Positions]) RM_idx = findall(x -> occursin("RM", x), Data[!, :Positions]) LWB_idx = findall(x -> occursin("LWB", x), Data[!, :Positions]) RWB_idx = findall(x -> occursin("RWB", x), Data[!, :Positions])
List of position vectors
position_vectors = [RW_idx, ST_idx, GK_idx, CM_idx, LW_idx, CDM_idx, LM_idx, CF_idx, CB_idx, CAM_idx, LB_idx, RB_idx, RM_idx, LWB_idx, RWB_idx]
Mutation function
function mutate(selected_players, position_vectors_list, probability) selected_players_matrix = copy(selected_players)
function select_random_player(idx_list, selected_players) while true random_idx = rand(idx_list) if random_idx ∉ selected_players return random_idx end end end for i in 1:nrows(selected_players) for pos_idx in 1:length(position_vectors_list) if rand() <= probability selected_players_matrix[i, pos_idx] = select_random_player(position_vectors_list[pos_idx], selected_players_matrix[i, :]) end end end return convert(DataFrame, selected_players_matrix)end
n_rows = 100 pop_init = convert(Matrix, DataFrame(Matrix{Union{Missing, Int}}(undef, n_rows, length(position_vectors)))) pop_init = fill!(pop_init, missing) pop_init = mutate(pop_init, position_vectors, probability=1)
Target function
function target(row, penalty=5) position_ratings = ["RWRating", "STRating", "GKRating", "CMRating", "LWRating", "CDMRating", "LMRating", "CFRating", "CBRating", "CAMRating", "LBRating", "RBRating", "RMRating", "LWBRating", "RWBRating"] parent_data = Data[row, :] ratings = parent_data[:, position_ratings] ratings_log = log.(ratings)
potential_minus_age = 0.15parent_data.Potential - 0.6parent_data.Age int_reputation = parent_data.IntReputation sumratings = sum(ratings_log) rating_list = vec(ratings_log) # Apply constraints constraint_penalty = 0 if sum(parent_data.ValueEUR) > 250000000 constraint_penalty += log((sum(parent_data.ValueEUR) - 250000000) ^ penalty) end if sum(parent_data.WageEUR) > 250000 constraint_penalty += log((sum(parent_data.WageEUR) - 250000) ^ penalty) end if any(rating_list .< 1.2) constraint_penalty += 1.2 ^ penalty end target_value = -(sumratings + 0.3sum(potential_minus_age) + sum(int_reputation)) + constraint_penalty return target_valueend
Tournament selection function
function tournament_selection(parents, t_size, penalty=6) random_parents_idx = sample(1:nrows(parents), t_size, replace=false) random_parents = parents[random_parents_idx, :]
random_parents_fitness = [target(parent, penalty=penalty) for parent in eachrow(random_parents)] best_parent_idx = argmin(random_parents_fitness) return random_parents[best_parent_idx, :]end
Crossover function
function crossover(parent1, parent2, crossover_point) offspring1 = vcat(parent1[1:crossover_point], parent2[(crossover_point + 1):end]) offspring2 = vcat(parent2[1:crossover_point], parent1[(crossover_point + 1):end])
return DataFrame([offspring1 offspring2]')end
Algorithm parameters
crossover_point = 7 population_size = 100 num_generations = 20000 tournament_size = 2 probability = 0.09 penalty = 1 parents = pop_init
global_best = pop_init[1, :] global_best_value = target(global_best)
Main loop
for gen in 1:num_generations # Parent population parent_pop = DataFrame(Matrix{Union{Missing, Int}}(undef, nrows(parents), length(position_vectors))) for c in 1:population_size parent_pop[c, :] = tournament_selection(parents, tournament_size, penalty=penalty) end parent_pop = fill!(parent_pop, missing)
# Generate offspring offspring_temp = DataFrame(Matrix{Union{Missing, Int}}(undef, 1, length(position_vectors))) for c in 1:2:population_size offsprings = crossover(parent_pop[c, :], parent_pop[c + 1, :], crossover_point) offspring_temp = vcat(offspring_temp, Matrix(offsprings)) end offspring_temp = offspring_temp[2:end, :] offspring_temp = fill!(offspring_temp, missing) parents = mutate(offspring_temp, position_vectors, probability=probability) # Evaluate solutions solutions = [target(parent) for parent in eachrow(parents)] idx_sol = argmin(solutions) temp_best = parents[idx_sol, :] temp_target_value = solutions[idx_sol] if temp_target_value <= global_best_value global_best = temp_best global_best_value = temp_target_value end penalty += 0.5 if penalty >= 4 penalty = 0 endend
global_best_fin = convert(Matrix, global_best') parent_data = Data[global_best_fin, :]
target(global_best)