MATLAB: What is Matlab doing when training a K-NN classifier

classification learnerknnStatistics and Machine Learning Toolbox

I am currently using the Classification Learner GUI to make a K-NN classifier model. Training (training data size: 800,000×52), is seemingly taking forever. I am wondering, what is Matlab doing while training? Since K-NN uses lazy learning, I thought there is no training required? So, why is it taking so long to "train"?

If this "training phase" is actually used to test unseen data to evalutate the model accuracy, is there any way to get more verbose output to see progress instead of just a indeterminate progress bar?

Best Answer

The training time might depend on the type of validation that you are using. If you are using k-fold validation, and the k is large it is likely that the model will take longer to train. The training phase uses the validation data (taken from the training dataset) to evaluate the accuracy of the model.

Related Solutions

MATLAB: Is the classifier trained twice when using holdout validation in ClassificationLearner App

The only purpose of the classifier that is trained on half of the dataset is to estimate the out-of-sample accuracy of the model that is trained on the full dataset.

The model trained on the full dataset is the one that is expected to perform better on out-of-sample data, because it is fit to more data. The key point here is that we care about the expected accuracy on out-of-sample data. Both classifiers are trained using the same hyperparameters (i.e., options like “number of learners”). Because they use the same hyperparameters, fitting to more data should only increase out-of-sample accuracy.

MATLAB: K-NN Classsification on images

Hi,

Your input arguments Training and Group must have the same number of rows. I see that Group has only 2 rows, but Training probably has many more rows that than. Training consists of tow matrices stacked on top of each other. These matrices represent images as M-by-N matrices. So if all your images (i1, i2, i3, and i4) are all of size M-by-N, then Sample and Training have 2M rows. If your intention is to treat each image as a single entity, then you could construct Sample and Training as follows:

Sample   = [i1(:) i2(:)]';
Training = [i3(:) i4(:)]';

That said, I don't think this classification approach is very useful if you only have one training example from each class.

-Arthur

Best Answer

Related Solutions

MATLAB: Is the classifier trained twice when using holdout validation in ClassificationLearner App

MATLAB: K-NN Classsification on images

Related Question