PyTorch is a Python package developed by Facebook AI designed to perform numerical calculations using tensor programming. It also allows its execution on GPU to speed up calculations.
Typically PyTorch is used both to replace numpy and process calculations on GPUs, and for research and development of machine learning algorithms including neural networks.
As part of the course at Jovian.ml “Deep Learning with PyTorch: Zero to GANs”, I’ll briefly explore five simple but useful functions of PyTorch package, providing some examples of a correct execution as well as cases in which it fails. The functions chosen are: