Setup

1. Clone this repository or download a .zip file and unzip to your local folder. Then navigate to a project's folder.

git clone https://github.com/v23ent/aa-hands-on.git && \
cd aa-hands-on

ℹ️ In case you have Jupyter already installed on your host and would like to avoid fiddling with Docker networking options, you can also just run jupyter notebook from the project folder and skip to step 4

2. Change you docker default network IP subnets by amending docker config file and restarting docker service: Open your docker daemon configuration file for editing. It should be located here:

• For Windows: C:\ProgramData\docker\config\daemon.json
• For Mac: ~/.docker/daemon.json
• Otherwise: /etc/docker/daemon.json

Add the following config lines:

  "default-address-pools" : [
    {
      "base" : "198.0.0.0/16",
      "size" : 24
    }
  ]

"base" could be different if you have the one mentioned above not available. The most important, it can't be 172.17.x.x :)

Save the file and restart your docker daemon:

• For Windows and Mac: restart your Docker Desktop service
• Otherwise:

sudo systemctl restart docker

3. Launch container with docker compose. Container would expose port 8888. If it is in use on your system, just use the other one amending docker-compose.yml with appropriate port.

docker compose up -d

4. Navigate with your to a dummy notebook and execute the first block of code. http://localhost:8888/notebooks/aa-demo-exercise.ipynb

It will install 'atscale' package with dependencies. If all goes well then you're ready go with AI-Link for Adaptive Analytics hands-on! We will be adding other blocks of code during hands-on session.

That's it!

Cleanup

1. Stop your containers

docker compose down

2. Delete added docker daemon config lines:

  "default-address-pools" : [
    {
      "base" : "198.0.0.0/16",
      "size" : 24
    }
  ]

3. Restart docker service on your machine in the same way as in Setup.