遍历R的ksvm中的多个C值

Question

Is there an easy way to iterate through multiple C values and display the top 5 results?

I have ksvm set up like this:

# call ksvm
model <-  ksvm(as.matrix(data[,1:10]),as.factor(data[,11]),type="C-svc",kernel="vanilladot",C=100, scaled=TRUE)
# calculate a1.am
a <- colSums(model@xmatrix[[1]] * model@coef[[1]])
a
# calculate a0
a0 <- -model@b
a0
# view predictions
pred <- predict(model,data[,1:10])
pred
# get model accuracy
sum(pred == data[,11]) / nrow(data) 

and want to know if there's an easy way to iterate through all values in a range for the C parameter (say 0.00001 - 10000000) and print only the top 5?

I'm new to R, but I think I want to do something like this:

allC = c(0.0001:10000000)
results=list()
for(i in 1:length(allC)){
  model <-  ksvm(as.matrix(data[,1:10]),as.factor(data[,11]),type="C-svc",kernel=vanilladot,C=allC[[i]],scaled=TRUE)
  results[[i]]=data.table(kernel=vanilladot,accuracy=sum(pred == data[,11]) / nrow(data))
}
rbindlist(results)

Answer 1

instead of allC = c(0.0001:10000000), you can make use of seq() function, see ?seq for details.

After that, assign rbindlist(results) to an object:

results_dt = rbindlist(results)
results_dt[order(-accuracy)][1:5] #this will give you top 5 w.r.t accuracy