Setting Up Wallaroo for Sample Pipelines and Models

How to get Wallaroo running for your sample models

Welcome to Wallaroo! We’re excited to help you get your ML models deployed as quickly as possible. These Quick Start Guides are meant to show simple but thorough steps in creating new pipelines and models. For full guides on using Wallaroo, see the Wallaroo SDK and the Wallaroo Operations Guide.

This first guide is just how to get Wallaroo ready for your first models and pipelines.

Prerequisites

This guide assumes that you’ve installed Wallaroo in a cluster cloud Kubernetes cluster.

Basic Wallaroo Configuration

The following will install the Wallaroo Quick Start samples into your Wallaroo Jupyter Hub environment.

  1. From the Wallaroo Quick Start Samples repository, visit the Releases page.

  2. Download the release that matches your version of Wallaroo. For example, the most recent release is torino. Either the .zip or the .tar.gz file can be downloaded. The example below uses the .zip file.

  3. Access your Jupyter Hub from your Wallaroo environment.

  4. Either drag and drop or use the Upload feature to upload the quick start guide .zip file.

  5. From the launcher, select Terminal.

  6. Unzip the files with unzip {ZIP FILE NAME}. For example, if the most recent release is 1.11-elise.zip, the command would be:

    unzip 11-elise.zip
    
  7. Exit the terminal shell.

The quick start samples along with models and data will be ready for use.