To learn how to use PyTorch, begin with our Getting Started Tutorials. The 60-minute blitz is the most common starting point, and provides a broad view into how to use PyTorch from the basics all the way into constructing deep neural networks.
Each tutorial is in the form of a Jupyter notebook. You can clone this project and run them in your Azure Notebooks compute.
Additional high-quality examples are available, including image classification, unsupervised learning, reinforcement learning, machine translation, and many other applications, in PyTorch Examples
You can find reference documentation for PyTorch's API and layers at PyTorch docs or via inline help.
Check out the PyTorch Cheat Sheet for additional useful information.