Using Codespaces to work with the "Practical Deep Learning for Coders" course

To get started, create a codespace for this repository by clicking this 👇

A codespace will open in a web-based version of Visual Studio Code.

Note: Dev containers is an open spec which is supported by GitHub Codespaces and other supporting tools.

Opening a notebook

The dev container is fully configured with software and machine learning libraries needed for this course.

In the VS Code editor, open any notebook file and start executing the notebook's cells.

Opening your codespace in JupyterLab

You can open your codespace in JupyterLab from the "Your codespaces" page at github.com/codespaces, or by using GitHub CLI with gh codespace jupyter. For more information, see "Opening an existing codespace".

GPU-powered Codespaces

GPU-powered Codespaces are now available in limited beta. Having access to a GPU from within a codespace allows developers to run complex Machine Learning models much more quickly.

To request access to the GPU machine types, or any additional machine type, please complete the sign up form.

Once, GPU is enabled and configured for your codespace, uncomment this section which installs NVIDIA CUDA.

Note: Notebooks 09-small-models-road-to-the-top-part-2 and 10-scaling-up-road-to-the-top-part-3 requires a powerful machine to ensure that the kernel does not crash. Hence, some notebook cells for these two notebooks might not execute without a GPU-powered codespace.