SkyPilot Tutorial 👩‍🏫

Welcome to SkyPilot tutorial! This tutorial is a collection of Jupyter notebooks that will walk you through the basics of using SkyPilot.

Installation

To setup the dependencies for this tutorial, run the following command:

pip install -r requirements.txt

Running the tutorial on your laptop

It is recommended to run the tutorial with jupyter lab. In a terminal in the root of the repository, run:

jupyter lab

Jupyter lab should now be running. You follow the URL generated by jupyter lab to open the tutorial in your browser.

If you do not have SkyPilot installed, please start with 00_installation or refer to the installation guide. Otherwise, you can start with 01_hello_sky.

Running the tutorial in a Docker container

If you prefer to run the tutorial in a Docker container, you can do so by running the following command:

docker run --rm -p 8888:8888 -it public.ecr.aws/a9w6z7w5/skypilot-tutorial:latest

Note that you may need to setup credentials inside the container.

If you would like to skip credential setup in the container and use your local credentials, copy your ~/.aws/ and ~/.config/ directories to a new directory /tmp/credentials such that it contains/tmp/credentials/.aws/ and /tmp/credentials/.config/gcloud/ directories with the relevant files. Then, run the following command:

docker run --rm -v /tmp/credentials:/credentials:ro 8888:8888 -it public.ecr.aws/a9w6z7w5/skypilot-tutorial:latest

This will automatically install the AWS and GCP credentials inside the container.