devcon_tune_SVM_libelinear.R

## DEVCON SVM (kernelab) TUNNING SCRIPT
## 27/04/2020

setwd("/home/harpo/Dropbox/ongoing-work/git-repos/devcon/phase3/")
#load("./deconv_cgdata_cps_less_feat.RData")
load("deconv_cgdata_cps.RData")
library(caret)
library(LiblineaR)
library(foreach)
library(doMC)
library(dplyr)
registerDoMC(cores=6)

epsilon_range <- c(0.005, 0.05, 0.1, 0.00005,0.0005)
cost_range <- c(0.0001, 0.001, 0.01, 0.1, 0.5, 1, 2, 4,8,16,32)

#epsilon_range <- c(0.005)
#cost_range <- c(0.0001)


parms <- expand.grid(cost = cost_range, epsilon = epsilon_range)
results_final <-c()
results_final_models <- list()
# Starting training  ---------
for (label_number in rownames(trainprop)) {
# Create datasets  ----------
  labels <- trainprop[label_number, ]
  trainset <- t(train)
  #trainset <- scale(trainset,center=TRUE,scale=TRUE)
  #data_train <- cbind(label = labels, trainset)
  
  labels_test <- testprop[label_number, ]
  testset <- t(test)
  #data_test <- cbind(label = labels_test, testset)
# Start parallel fine tuning  -----------
  results <- foreach(i = 1:nrow(parms), .combine = rbind) %dopar% {
    c <- parms[i,]$cost
    e <- parms[i,]$epsilon
    model <- LiblineaR(
      target=labels,
      data = trainset,
      type = 12,
      cost = c,
      svr_eps = e,
    )
    
    preds <- predict(model, testset)
    spear <- cor(x = preds$predictions, y = labels_test, method = "spearman")
    pears <- cor(x = preds$predictions, y = labels_test, method = "pearson")
    partial_results<-data.frame(label_number,parms[i,], pearson = pears, spearman = spear)
    partial_results 
    
  }
# Select best model ---------
  best_model <- results %>% arrange(desc(pearson)) %>% filter(row_number()==1) 
  print(paste("selecting best model for ",best_model$label_number," : ",best_model$cost, ", ", best_model$epsilon," Pearson value : ", best_model$pearson,sep=""))
  model <- LiblineaR(
    target=labels,
    data = trainset,
    type = 12,
    cost = best_model$cost,
    svr_eps = best_model$epsilon,
  )
  results_final_models[[label_number]]<-model
  save(results_final_models,file = "results_liblinear_devcon_bestmodels_fgdata_cps_12_noscale.rdata",compress = "gzip")
  results_final<-rbind(results_final,results)
  readr::write_csv(results_final,path="results_liblinear_devcon_fgdata_cps_12_noscale.csv")
}