This lab provides a hands-on experience with RHEL AI and InstructLab, guiding you through the process of understanding, training, and improving Large Language Models.
In order to test and develop on your local machine, you can use a specially built container with Podman or Docker as follows.
-
Create a git repo from this template
-
Suggested naming:
showroom_<lab-name>
-
-
Clone your new repo and
cd
into it -
When you make changes to the content, all you need is to kill the container and run it again.
podman run --rm --name antora -v $PWD:/antora -p 8080:8080 -i -t ghcr.io/juliaaano/antora-viewer
Live-reload is not supported.
-
Create a git repo from this template
-
Clone the repo and
cd
into it -
Run ./utilities/lab-serve
-
Open http://localhost:8080 in your browser
-
Run ./utilities/lab-build to build your html
To rebuild your html, run ./utilites/build
.
Now you are ready to go!
You can start editing the files in the content/modules/ROOT/pages/
directory.
Many modern editors such as Visual Studio Code offer live Asciidoc Preview extensions.
./content/modules/ROOT/
├── assets
│ └── images # Images used in your content
│ └── example-image.png
├── examples # You can add downloadable assets here
│ └── example-bash-script.sh # e.g. an example bash script
├── nav.adoc # Navigation for your lab
├── pages # Your content goes here
│ ├── index.adoc # First page of your lab, e.g. overview etc
│ ├── module-01.adoc
│ └── module-02.adoc # Sample lab has 2 modules including index.adoc
└── partials # You can add partials here, reusable content inserted inline into your modules
└── example_partial.adoc
As a convenience to developers, the Dev Mode Extention (disabled by default) displays the asciidoc attributes you have to work with while writing your lab instructions.