Hi there!
I was wondering how I can develop a Machine Learning algorithm that has as a train dataset some images (e.g. photos of animals associated with the species) and as a test dataset other images (e.g. photos of other animals to be assigned the species class).
I can develop an algorithm from a dataset composed of data, but I don't know where to start having images. Could someone give me some suggestions?
One of the best introductions I've seen to building neural networks (including image models) is the book "Deep Learning with R". If you're new to neural networks I would suggest starting there.