Arundo Composer

Arundo Composer gives you the ability to publish machine learning models into the Arundo Fabric cloud. This lets you add models to pipelines or deploy them to Arundo Edge so you can run data through the models and enhance your data analytics.

Composer matrix


Key Features

End-to-end native data science workflow

Allows data science teams to use their preferred tools to build and deploy their models accessing the full suite of tools they need to build and share predictive models on data sets residing in the cloud. Native data science languages such as Python, R, and Scala supported.

One-click deploy

Quickly publish models as applications that run in a web browser, allowing analysts and engineers to test, interact, and deploy as required.


Components

Arundo Composer is made up of two components.

Arundo Composer CLI

Command-line interface that interacts with your data science environment. This interaction gives you the ability to:

  • Deploy models for testing
  • Publish models into the Arundo Fabric cloud

Arundo Composer Runtime

Package of modules that builds REST endpoints for models. This gives you the ability to:

  • Test models in a model application (web browser-based application)
  • Interact with models using third-party applications

How it Works

Here's how to use Arundo Composer.

Step 1: Build

Use the Arundo Composer CLI to build a model workspace.

Tip

A model workspace contains the app.py and config.yaml files necessary to test and publish a model.

Then, build the model itself by updating the app.py and config.yaml files to your specifications.

Step 2: Test

Next, use the Arundo Composer CLI to run the model. The Arundo Composer Runtime then builds REST endpoints for the model. This gives you the ability to:

  • Test the model in a model application (web browser-based application)
  • Interact with the model using third-party applications

Step 3: Publish

Finally, use the Arundo Composer CLI to publish the model to Arundo Fabric. From Arundo Fabric, you can:

  • Connect the model to pipelines
  • Link pipelines featuring the model to streaming data from industrial assets
  • View visualizations of the model's output on dashboards
  • Deploy the model to Arundo Edge so you can run analytics on the edge