How can I see the predicted labels for all my validation set images? I used deep network designer to modify a pre-trained network .

Hi,

So I used Deep Network Designer ( DND) to modify GoogleNet and do some transfer learning on a new dataset. I separated the data into training vs validation, again using the DND. When the transfer learning process has ended, I obtained an accuracy of 97. Now, my issue is the deployment of the results. I need a code that gives me a file with all the images and the predcited class and its percentage.

For the moment I only have this

idx = randperm(numel(imdsValidation.Files),4);

figure

for i = 1:4

subplot(2,2,i)

I = readimage(imdsValidation,idx(i));

imshow(I)

label = YPred(idx(i));

title(string(label) + ", " + num2str(100*max(probs(idx(i),:)),3) + "%");

end

But obsly if I adapted it for hundreds of images does not work.

Please let us know if there is any possibility to deploy all the validation images with the class predicted.

Thanks a lot,

Andreea

Answer 1

[Prince Kumar]

You can pass all the validation set as a batch to get the all predictions at once instead of doing it one by one.

For example:

[YPred,probs] = classify(net,imdsValidation);

Here net can be any network which you trained.

For a full example you can have a look at this link :

https://www.mathworks.com/help/deeplearning/ref/trainnetwork.html