clc; close all; clear;
Tic.

% % load modelLoad (' - mat ', 'C: \ Users \ \ Desktop \ convolution case 001. The mat');
% % loading test set
Location='E: \ image_test \ test_02';
Imds=imageDatastore (" C: \ Users \ Desktop \ training \ ', 'IncludeSubfolders, true,' LabelSource ', 'foldernames);;
InputSize=lgraph. The Layers (1). InputSize;
Imdstest=augmentedImageDatastore (inputSize (1:2), imds);
% tic;
[YPred, scores]=the classify (lgraph imdstest);

% disp ([' time is used for the classification: ', num2str (toc), 'seconds']);

I find on the net migration procedure classify function to learn the first parameter is trained network model, the second parameter is the test set, but I'll help document
The classify function to wear at least three parameter
Syntax
Class=the classify (sample, training, group)
Class=the classify (sample, training, group, 'type')
Class=the classify (sample, training, group, 'type', the prior)
[class, err]=the classify (... )
[class, err, POSTERIOR]=the classify (... )
[class, err, POSTERIOR, logp]=the classify (... )
[class, err, POSTERIOR, logp, coeff]=the classify (... )
Hope everybody who help me to answer it thank you very much!!!!!!!!!!