Help with Leave One Out Cross Validation

Hello everyone!

I'm in a very large pickle right now, as it is my exam period, and we have to create a MatLab document wherein we have to build a lot of classifiers and those kinds of things.

I have a dataset (found here (<http://archive.ics.uci.edu/ml/datasets/seeds#))>, and I need to perform a Leave One Out Cross Validation on it, but I'm at a loss of how to even get started.

Can anyone give me some tips on how to get started? I'd really appreciate it.

Thank you guys,

-Mads

Answer 1

[Sean de Wolski]

I would start with crossval.

doc crossval

It looks like crossval even has a builtin in option for 'leaveout' which does this.

Answer 2

[Mads]

Thank you for the quick reply.

It says that crossval is a "Loss estimate using cross-validation" - what does "loss estimate" mean?

I've also found cvpartition, which also has an option for leave-out, but I'm unsure of which data to feed into the function, or even what I'm supposed to get out of it.

Thanks for taking the time.

-Mads

[CHHAVI]

How you did cross validation/ leave one out?. please help