Emily's Cat

I'm not interesting enough for anyone to give a crap about my status...

Happy birthday Emily!

So earlier today I decided I was going to use the best.svm function in the e1071 package of R to try the values seq(0, 0.1, by=0.01) for gamma and c(1, 5, 10, 20) for cost to determine which pair would result in optimal performance on my test set and it's running now but it's going to take a while with the default rbf kernel, which I want—I'm afraid a faster linear kernel wouldn't capture the relationships between the attributes accurately. Being as you are a highly qualified statistician and actuary far ahead of any mere computer programmer in the area of machine learning, as you claim, I was wondering if you had any pointers that could steer me in the direction of a faster way to get the desired outcome. This is the dataset I'm working with:

https://www.kaggle.com/harlfoxem/housesalesprediction

Humbly yours,
Prokhor