Low-Code ML with Azure Machine Learning

Kevin Feasel (@feaselkl)
https://csmore.info/on/lowcodeaml

Who Am I? What Am I Doing Here?

Catallaxy Services
Curated SQL
Finding Ghosts in Your Data

@feaselkl

Motivation

My goals in this talk:

  • Provide an overview of Azure Machine Learning.
  • Introduce Automated Machine Learning (AutoML).
  • Train an ML model through the designer.
  • Deploy an ML model and score data, individually and in bulk.

What We'll Do

What We'll Do

What We'll Do

What We'll Do

What We'll Do

Agenda

  1. A Brief Primer on Azure ML
  2. Automated Machine Learning (AutoML)
  3. Training a Model
  4. Deploying a Model

What is Azure Machine Learning?

Azure Machine Learning is Microsoft's primary offering for machine learning in the cloud.

Starting Out: Workspaces

A workspace is necessary to get started with Azure ML.

Workspace Basics

Advanced Workspace Settings

Key Components

There are several major components which make up Azure ML.

  • Datastores
  • Datasets
  • Compute instances
  • Compute clusters
  • Azure ML Studio Designer
  • Experiments and Runs
  • Models
  • Endpoints
  • Inference clusters

Datastores

Datastores are connections to where the data lives, such as Azure SQL Database or Azure Data Lake Storage Gen2.

Datasets

Datasets contain the data we use to train models.

Compute instances

Compute instances are hosted virtual machines which contain a number of data science and machine learning libraries pre-installed. You can use these for easy remote development.

Compute clusters

Sometimes, you want something a bit more powerful to perform training. This is where compute clusters can help: spin them up for training and let them disappear automatically afterward to save money.

Designer

The Azure ML Studio Designer allows you to create training and scoring pipelines using a drag-and-drop interface reminiscent of SQL Server Integration Services.

Experiments and Runs

Experiments allow you to try things out in a controlled manner. Each Run of an experiment is tracked separately in the experiment, letting you see how well your changes work over time.

Models

The primary purpose of an experiment is to train a model.

Endpoints

Once you have a trained model, you can expose it as an API endpoint for scoring new data.

Inference clusters

Inference clusters are an easy method to host endpoints for real-time or batch scoring.

Demo Time

Agenda

  1. A Brief Primer on Azure ML
  2. Automated Machine Learning (AutoML)
  3. Training a Model
  4. Deploying a Model

What is AutoML?

Automated Machine Learning (AutoML) is a process intended to simplify the machine learning process. It provides two benefits:

  1. It reduces the level of skill needed to run a machine learning project.
  2. It provides us a minimum baseline of model quality--skilled practitioners can beat AutoML but it provides a nice "first approach" score.

Performing AutoML

Select a Dataset

Configure Model Parameters

Configure Features

Configure Features

Configure Features

Configure Features

Review Results

Demo Time

Agenda

  1. A Brief Primer on Azure ML
  2. Automated Machine Learning (AutoML)
  3. Training a Model
  4. Deploying a Model

Using the Designer

The Azure Machine Learning process:

  • Get some data
  • Clean up that data
  • Choose an algorithm
  • Train the model
  • Evaluate the model

Demo Time

Agenda

  1. A Brief Primer on Azure ML
  2. Automated Machine Learning (AutoML)
  3. Training a Model
  4. Deploying a Model

Deploying a Model

Once we have trained a model, our next goal is to make it available.

Build a Pipeline

The first step is to build an inference pipeline. This is different from a training pipeline: we don't use the training dataset; instead, we accept user input and return a scored response.

Two Approaches

There are two approaches to inference pipelines:

  1. Real-time inference -- best for generating a few predictions
  2. Batch inference -- best for scoring large numbers of records

Let's look at real-time inference today.

Demo Time

Wrapping Up

Over the course of this talk, we have gained an understanding of the basics of low-code machine learning with Azure ML. This includes using the UI-heavy pipeline designers as well as taking advantage of AutoML.

Wrapping Up

To learn more, go here:
https://csmore.info/on/lowcodeaml

The next talk in the series:
https://csmore.info/on/amlindepth

And for help, contact me:
feasel@catallaxyservices.com | @feaselkl


Catallaxy Services consulting:
https://CSmore.info/on/contact