First Project on ML.NET – Beginner Guide

Machine Learning is becoming very popular, and if you are a .NET developer, the best way to start Machine Learning is by using ML.NET. In this blog, we will create our first Machine Learning project using ML.NET step-by-step in simple language.

This guide is perfect for beginners who know C# / .NET but are new to Machine Learning.

1. What is ML.NET?

ML.NET is a Machine Learning framework created by Microsoft that allows .NET developers to build AI/ML models using C# or F# without learning Python.

With ML.NET you can build:

Spam detection
Price prediction
Recommendation system
Sentiment analysis
Image classification

Official site:
https://dotnet.microsoft.com/apps/machinelearning-ai/ml-dotnet

2. System Requirements

Before starting, install:

Visual Studio 2022
.NET SDK 6 or later
NuGet Package Manager

Install ML.NET package:

Install-Package Microsoft.ML

3. Our First Project – Price Prediction Model

We will create a simple ML model that predicts House Price based on size.

Example data:

Size	Price
1	10000
2	20000
3	30000
4	40000

We will train ML model using this data.

4. Step 1 – Create Console Project

Create new project:

Console App (.NET)

Install package:

Microsoft.ML

5. Step 2 – Create Data Model

Create class:

using Microsoft.ML.Data;

public class HouseData
{
    [LoadColumn(0)]
    public float Size;

    [LoadColumn(1)]
    public float Price;
}

public class HousePrediction
{
    [ColumnName("Score")]
    public float Price;
}

6. Step 3 – Write ML Code

using Microsoft.ML;
using System;
using System.Collections.Generic;

class Program
{
    static void Main()
    {
        var context = new MLContext();

        var data = new List<HouseData>()
        {
            new HouseData{ Size = 1, Price = 10000 },
            new HouseData{ Size = 2, Price = 20000 },
            new HouseData{ Size = 3, Price = 30000 },
            new HouseData{ Size = 4, Price = 40000 }
        };

        var trainingData = context.Data.LoadFromEnumerable(data);

        var pipeline = context.Transforms
            .Concatenate("Features", nameof(HouseData.Size))
            .Append(context.Regression.Trainers.Sdca(
                labelColumnName: "Price",
                maximumNumberOfIterations: 100));

        var model = pipeline.Fit(trainingData);

        var predictor = context.Model.CreatePredictionEngine
            <HouseData, HousePrediction>(model);

        var prediction = predictor.Predict(
            new HouseData { Size = 5 });

        Console.WriteLine(
            $"Predicted price: {prediction.Price}");
    }
}

7. Output

Predicted price: 50000 (approx)

Model learned pattern:

Price = Size × 10000

This is called Regression Model in Machine Learning.

8. What We Learned

In this first ML.NET project we learned:

What is ML.NET
How to install ML.NET
How to create model
How to train data
How to predict value

This is the basic workflow of Machine Learning.

Data → Train → Model → Predict

Conclusion

ML.NET makes Machine Learning easy for .NET developers.
You do not need Python to start AI.
With ML.NET you can build real AI applications using C#.