Greetings to all,
i am working on camera based, script identification there i am extracting the features from the document image and use classifiers to obtain the identification result. for this purpose i am using K-NN classification there i am using k-fold cross validation of 2-fold, 5-fold and 10-fold. can anybody explain these three cross validations. how the data is split in 2-fold, 5-fold and in 10-fold cross validation.
thanking you,
Regards.
SATISH KUMAR
Dear Satish,
k-fold cross validation splits the dataset in k different (disjoint) subsets of approximately the same dimension, and use in turn one of the subsets for estimating the generalization error, and the others for trainng the system (in your case a K-NN classifier).
If you are performing 2-fold cross validation, you are dividing your dataset into two halves, and first you use the first one for training and the second for testing, and then you use the second for training and the first for testing. Averaging the testing error over two realizations is not very significant though ...
When you are performing 5-fold cross validation, you make 5 rounds of the training-testing procedure, each time using one different subset (out of the 5 available) for testing and the other 4 for training.
Hope this helps
Enrico
I recommend 100-fold cross validation because there are many merits.
See my book " New Theory of Discriminant Analysis after R. Fisher (Springer)". There are many examinations.
- Badges
- Science topic
- Similar topics
- Condensed Matter Physics
- Zero