Where is the problem with this code?

1 view (last 30 days)
lech king
lech king on 15 Apr 2021
Answered: Cris LaPierre on 15 Apr 2021
I will receive this message after entering the information(dsnew) into the deep learning application
invalid training data for classification network response must be categorical
location1 = fullfile(matlabroot,'bin','F18','test9','noise');
location2 = fullfile(matlabroot,'bin','F18','test9','1','main');
location3 = fullfile(matlabroot,'bin','F18','test9','1','validation');
noise = imageDatastore({location1},'FileExtensions',{'.jpg','.png','.jpeg'},'IncludeSubfolders',true,'LabelSource','foldernames');
nonnoise = imageDatastore({location2},'FileExtensions',{'.jpg','.png','.jpeg'},'IncludeSubfolders',true,'LabelSource','foldernames');
validation = imageDatastore({location3},'FileExtensions',{'.jpg','.png','.jpeg'},'IncludeSubfolders',true,'LabelSource','foldernames');
aug1 = imageDataAugmenter('RandRotation',[0 90],'RandScale',[1.1 1.3]);
auimds1 = augmentedImageDatastore([224 224 1],nonnoise,'ColorPreprocessing','rgb2gray','DataAugmentation',aug1);
auimds2 = augmentedImageDatastore([224 224 1],noise,'ColorPreprocessing','rgb2gray');
validation1 = augmentedImageDatastore([224 224 1],validation,'ColorPreprocessing','rgb2gray');
dsnew = combine(noise,nonnoise);
  2 Comments
Jan
Jan on 15 Apr 2021
It is a bad idea to store data in Matlab bin directory.
Please post the complete error message. This is better than letting the readers guess, which lines causes the problem.
lech king
lech king on 15 Apr 2021
It is true that my explanations were not enough
I'm sorry
The code executes correctly
I have merged 2 training data, one with noise and the other without noise, in the last line and saved in the dsnew variable
The deep learning app show this message when running

Sign in to comment.

Answers (1)

Cris LaPierre
Cris LaPierre on 15 Apr 2021
Inspect your response variable. It apparently has the wrong data type.
If you don't yet know how to create categorical data, see this link.

Products


Release

R2020b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!