In this four-part tutorial, you’ll learn all the basics about torch, all while playing with a neural network from the outset!
Right at the start, you’ll build a neural network from scratch, using nothing but functionality inherent in torch’s basic data structure: tensors. In the remaining three tutorials, you’ll successively get rid of all low-level operations and let torch do all the hard work for you:
first, by letting it do automatic differentiation;
second, by using neural network modules, instead of bare tensors; and
third, by calling torch optimizers to update the model’s weights.
Once you’re done, you’re more than ready to go on and explore some of the applied examples!