'How do I find the 20 nearest neighbors for the first 100 instances to the mean of each class in this 256x1100x10 uint8 dataset using Matlab?
This is the code I have so far. I want to use the Matlab function fitcknn to accomplish this but don't know where to start
clear, clc
load('usps_all.mat')
averaged_by_class_data=mean(data,2) %mean of each class - there are 10 classes
for class=1:10
first_100_images=data(:,1:100,class)
Mdl = fitcknn(averaged_by_class_data(:,:,class),first_100_images,'NumNeighbors',20,'Standardize',1)
end
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|
