Creating a object recognition dataset for deep learning in MATLAB

I've been working through Andrew Ng's Deep Learning tutorials and would like to apply a Sparse Autoencoder to my own data set.

I have begun collecting and categorizing images but I have a question:

How exactly does one go from a collection of images in a folder to a properly formatted dataset to be used in MATLAB? An example of such a dataset is the MATLAB formatted STL-10 dataset .

For this problem, the inputs are grey scale images of toys split into 5 categories (this is a classification problem)

Thanks

[Avi Nehemiah]

The best way to go from a collection of images to a properly formatted dataset is to use the imageDatastore.

This example shows how this can be done for a classification problem ( http://www.mathworks.com/help/vision/examples/image-category-classification-using-deep-learning.html )