ZBL Command Line Interface

Install and Configure ZBL-cli in a Jupyter notebook

Spawn a Jupyterlab notebook and open it in the browser

Open the Terminal and perform the procedures that are mentioned:

Install the Zeblok CLI into the notebook using the command below:

npm i -g zbl-cli-modular

Zeblok CLI is used in the workstation to interact with AI-Microcloud for model serving, pipeline, and snapshot

Configure the command line interface with the Zeblok account

zeblok configure <platformUrl> <datalake url> <datalakeSecret> <username>

Platform URL: Please refer to the image attached

Datalake URL: Please contact the Zeblok Admin for this information.

Datalake Secret: Select the Kebab menu and go to the "Manage your Zeblok Account", then click on the "Get Secret Key" Button and you can copy the alphanumeric key under the "Ai-DataLake Secret Key"

Username: You can copy your user name from the same page you got your Datalake Secret from. Please follow the picture below

When the workstation is spawned, this command is pre-configured into your notebook with all configurations

Authentication of the Microcloud with the CLI

zeblok auth

Once you enter the above command the cli will prompt for Access Key and Secret Key

You can get both the credentials from the "Manage your Zeblok Account" option. Click on the "Get API Credential" button to reveal both keys, and you can copy and paste them from there.

After you have entered both the keys into the CLI after then you can easily use the CLI to build your own Models or to serve pre-installed models using the Zeblok command line.

Run the command below to check if the ZBL-CLI is working

zeblok

Once the initiation of any services like model serving is initiated try not to regenerate the access key and secret key again in profile tab to avoid failure in updates to app .