#Include library
library(e1071)
library(mvtnorm)
library(fields)

#Reading .csv
inp <- read.csv('simclass1_train.csv',sep=';')

#Making a data frame
X <- inp[,-1]
Y <- inp[,1]
dat <- data.frame(Y=as.factor(Y),X=X[,c(2,1)])

#Plotting
par(mar=c(3.1,3.1,1.6,1.5),mgp=c(1.7,0.6,0),font.main=1,cex.main=0.8)
plot(X,col=Y+1,xlab='X1',ylab='X2',asp=1)

## Fitting SVM
set.seed (1)
tune.out <- tune(svm,Y~., data=dat, kernel="radial",
                 ranges=list(cost=c(1),
                             gamma=c(100)))
best.svm <- tune.out$best.model

summary (tune.out)

#Plotting boundary for SVM
xgrid <- seq(from=-6,to=6,by=0.1)
ygrid <- seq(from=-6,to=6,by=0.1)
xygrid <- expand.grid(xgrid,ygrid)
colnames(xygrid) <- c('X.X1','X.X2')
pgrid <- as.numeric(predict(best.svm,newdata=xygrid))-1

imp <- as.image(Z=pgrid,ind=xygrid,nx=100,ny=100)
image(imp,col = two.colors(start='blue',end='red',alpha=0.3),xlim=range(X[,1])+c(-1,1),ylim=range(X[,2])+c(-1,1),zlim=c(0,1))
contour(imp,levels=0.1,add=TRUE,lty='dashed')
points(X,col=Y+1)