Search⌘ K
Home
Courses
Using ML.NET to Build Machine Learning Models

Building an AutoML Pipeline

Explore building an AutoML pipeline in ML.NET to automate machine learning tasks for regression problems. Understand the process of loading data, configuring experiments, and evaluating models to identify the best fit. This lesson helps you gain hands-on knowledge of setting up AutoML for efficient model training and feature importance analysis.

We'll cover the following...

In this lesson, we'll look at building a basic AutoML pipeline in ML.NET. We'll do so with the aid of the following playground. This project represents an AutoML pipeline for a regression task that builds a model to predict house prices.

<Project Sdk="Microsoft.NET.Sdk">

  <PropertyGroup>
    <OutputType>Exe</OutputType>
    <TargetFramework>net6.0</TargetFramework>
    <PlatformTarget>x64</PlatformTarget>
    <ImplicitUsings>enable</ImplicitUsings>
    <Nullable>enable</Nullable>
  </PropertyGroup>

  <ItemGroup>
    <PackageReference Include="Microsoft.ML.AutoML" Version="0.20.1" />
  </ItemGroup>

  <ItemGroup>
    <None Update="lisbon-house-prices.csv">
      <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
    </None>
  </ItemGroup>

</Project>
Basic AutoML pipeline demo

AutoML dependencies

To run AutoML inside ML.NET, we'll need to reference the Microsoft.ML.AutoML NuGet package. We do so in line 12 of the MLApp.csproj file.

For our trial run, we use the lisbon-house-prices.csv file as the dataset. This file represents house prices in Lisbon, Portugal. It has the following columns:

  • condition ...