What type of machine learning are you doing? Cross validation is fairly straightforward, all you need to do is use some random part of your data for training and other part of the data for testing and you may do this several times. A popular method is called leave-one-out. Where you divide your data into several chunks and use all except one to train and the last one to test.
There is no special code required, if you have your training and testing methods ready, its just about how you partition and provide data to these function.
Take a look at the other common types of cross validation methods:
Of course if you have Stats toolbox, they have a function that can take your training function as input and itself splits the data for you, but its not really that much of an effort to do it.
