Hello Carl, The datasets used to train such networks are not available on that repository, partly because of the huge data size and copyright issues. You can access some smaller datasets available at the data repository. You can also download the ImageNet yourself at their website. As for how to train them, you can check the documentation on NetTrain. There's a section on out-of-core learning, also available as a tutorial here. If the data are images, you just pass a File dataset and Mathematica handles the rest. Regards, Lombardo

Hi Lombardo, Thanks for your response. I understand that the datasets and so on are very large - I'm not really looking for the exact training code, but an example of how to train this specific network (a simple example of one image -> output, for instance). Since some of the networks may use specific methods (ADAM, etc), it's important to know this if I want to train the network on my own data. So to clarify, I'm not looking for the exact training code, because I know that might be huge (or for all I know, not done in Mathematica). What I really need is a simple example invocation of NetTrain for the networks. Cheers, Carl

And let me just say, the neural net repo is one of the most useful things I've seen in years. Well done.

Can examples for training be shown in the Neural Network repository?