How can I apply PCA on thousand of images to reduce the dimension of the data?

3 vues (au cours des 30 derniers jours)

MUHAMMAD ALKHUDAYDI le 29 Août 2019

0
Lien

Utiliser le lien direct vers cette question

https://fr.mathworks.com/matlabcentral/answers/478168-how-can-i-apply-pca-on-thousand-of-images-to-reduce-the-dimension-of-the-data

Commenté : Subhadeep Koley le 6 Sep 2019

Hi,

I applied a simple CNN on dataset of 1000 images and the code works fine. However, I want to apply PCA on both training and testing set in order to reduce reduce its dimensionality (features selection) the code is below. Many thanks.

close all
clear all 
clc
%% Load image data
imds = imageDatastore('the path of images folder',...
    'IncludeSubfolders',true,'LabelSource','foldernames')   
tbl = countEachLabel(imds)
%%  Divide the data into 70% training data and 30% validation data
[imdsTrain,imdsValidation] = splitEachLabel(imds,0.7,'randomized');
%% Specify training options.
options = trainingOptions('sgdm', ...
    'MiniBatchSize',10, ...
    'MaxEpochs',3, ...
    'Shuffle','every-epoch', ...
    'InitialLearnRate',1e-4, ...
    'ValidationData',imdsValidation, ...
    'ValidationFrequency',6, ...
    'Verbose',false, ...
    'Plots','training-progress', ...
    'ExecutionEnvironment','gpu');
%% Layers for simple CNN.
layers_1 = [
    imageInputLayer([512 512 1])
    
    convolution2dLayer(3,8,'Padding','same')
    batchNormalizationLayer
    reluLayer
    
    maxPooling2dLayer(2,'Stride',2)
    
    convolution2dLayer(3,16,'Padding','same')
    batchNormalizationLayer
    reluLayer
    
    maxPooling2dLayer(2,'Stride',2)
    
    convolution2dLayer(3,32,'Padding','same')
    batchNormalizationLayer
    reluLayer
    fullyConnectedLayer(2)
    softmaxLayer
    classificationLayer];
%% train the network
netTransfer = trainNetwork(imdsTrain,layers_1,options);