How to develop a mini-batch datastore of images?

Hello,

I would like to see an example of defining a custom mini-batch datastore that uses images (for the purpose of deep learning). Labels of images are subfolder names.

Thanks!

Answer 1

[yanqi liu]

yes，sir，may be consider view 《Create Simple Deep Learning Network for Classification》

https://www.mathworks.com/help/releases/R2019a/deeplearning/examples/create-simple-deep-learning-network-for-classification.html

the data with subfoler can get by

https://www.mathworks.com/help/releases/R2019a/deeplearning/examples/train-deep-learning-network-to-classify-new-images.html

unzip('MerchData.zip');

imds = imageDatastore('MerchData', ...

'IncludeSubfolders',true, ...

'LabelSource','foldernames');

[imdsTrain,imdsValidation] = splitEachLabel(imds,0.7);

and set trainingOptions by MiniBatchSize can make mini-batch parameter