Example of a docker image usage

I would like to see some examples of how I might use a good PyTorch in a Docker container?

What sorts of examples are you looking for?

The fastest way to run pytorch in a docker image is to run the following command:
nvidia-docker run --rm -ti --ipc=host pytorch/pytorch:latest

assuming you have Docker and nvidia-docker set up.

I am running on a MAC so nvidia-docker is not an option. I was looking to process a python script or Jupyter notebook. Examples of what volumes are available as well as ports to interact with the image?

I think you should switch from macOS to any linux of your choice if you want to use pytorch in a docker, because docker on a mac runs in a kind of virtual machine with no direct access to the gpus. You can install pytorch normaly, but on a mac the nvidia driver is not well supported. It is much easier to use pytorch on a linux system.