knnclassify to fitcknn conversion

how to i convert this line of code to fitcknn?

knnclassify(test (:,1:27), train (:,1:27), train (:,27+1),k);

test and train are xls files

k = 3

Answer 1

[Ishu]

Hi Dafni,

As you want to train your model for "k-nearest neighbor classsifier" using 'fircknn()' you can follow these steps:

First you have to load your data from the Excel files using 'xlsread'.

trainData = xlsread('train.xls');
testData = xlsread('test.xls');

Next you can sepearte the features according to the line of code you have provided.

train = trainData(:, 1:27);
trainLabels = trainData(:, 28);
test = testData(:, 1:27);

Now you can train the k-nearest neighbours classifier using 'fitcknn()' with specified value of 'k'. In your case value of 'k' is 3.

knnClassifier = fitcknn(train, trainLabels, 'NumNeighbors', k);

Further you can use 'predict()' to predict for the 'test'.

predictedLabels = predict(knnClassifier, test);

For more information on 'fitcknn()', you can refer this documentation:

https://www.mathworks.com/help/stats/fitcknn.html

Hope it helps!