[C#/ML.NET] 선형 회귀 모델 사용하기

TestProject.zip

다운로드

▶ HouseData.cs

namespace TestProject
{
    /// <summary>
    /// 주택 데이터
    /// </summary>
    public class HouseData
    {
        //////////////////////////////////////////////////////////////////////////////////////////////////// Property
        ////////////////////////////////////////////////////////////////////////////////////////// Public

        #region 크기 - Size

        /// <summary>
        /// 크기
        /// </summary>
        public float Size { get; set; }

        #endregion
        #region 가격 - Price

        /// <summary>
        /// 가격
        /// </summary>
        public float Price { get; set; }

        #endregion
    }
}

 

728x90

 

▶ Prediction.cs

using Microsoft.ML.Data;

namespace TestProject
{
    /// <summary>
    /// 예측
    /// </summary>
    public class Prediction
    {
        //////////////////////////////////////////////////////////////////////////////////////////////////// Property
        ////////////////////////////////////////////////////////////////////////////////////////// Public

        #region 가격 - Price

        /// <summary>
        /// 가격
        /// </summary>
        [ColumnName("Score")]
        public float Price { get; set; }

        #endregion
    }
}

 

300x250

 

▶ Program.cs

using System;
using System.Globalization;
using System.Threading;

using Microsoft.ML;

namespace TestProject
{
    /// <summary>
    /// 프로그램
    /// </summary>
    class Program
    {
        //////////////////////////////////////////////////////////////////////////////////////////////////// Method
        ////////////////////////////////////////////////////////////////////////////////////////// Static
        //////////////////////////////////////////////////////////////////////////////// Private

        #region 프로그램 시작하기 - Main()

        /// <summary>
        /// 프로그램 시작하기
        /// </summary>
        private static void Main()
        {
            Console.WriteLine("BEGIN MAIN FUNCTION");

            CultureInfo cultureInfo = new CultureInfo("en-us");

            Thread.CurrentThread.CurrentCulture   = cultureInfo;
            Thread.CurrentThread.CurrentUICulture = cultureInfo;

            MLContext context = new MLContext();
   
            Console.WriteLine("BEGIN SET TRAINING DATA VIEW");

            HouseData[] trainingArray =
            {
                new HouseData() { Size = 1.1f, Price = 1.2f },
                new HouseData() { Size = 1.9f, Price = 2.3f },
                new HouseData() { Size = 2.8f, Price = 3.0f },
                new HouseData() { Size = 3.4f, Price = 3.7f }
            };

            IDataView trainingDataView = context.Data.LoadFromEnumerable(trainingArray);

            Console.WriteLine("END SET TRAINING DATA VIEW");

            Console.WriteLine("BEGIN SET PIPELINE");

            var pipeline = context.Transforms.Concatenate("Features", new[] { "Size" })
                .Append(context.Regression.Trainers.Sdca(labelColumnName : "Price", maximumNumberOfIterations : 1000));
   
            Console.WriteLine("END SET PIPELINE");

            Console.WriteLine("BEGIN SET MODEL");

            var model = pipeline.Fit(trainingDataView);
   
            Console.WriteLine("END SET MODEL");

            Console.WriteLine("BEGIN PREDICT");

            HouseData houseData = new HouseData() { Size = 2.5f };

            Prediction predition = context.Model.CreatePredictionEngine<HouseData, Prediction>(model).Predict(houseData);

            Console.WriteLine("--------------------------------------------------");
            Console.WriteLine($"SIZE  : {houseData.Size * 1000} SQ FT");
            Console.WriteLine($"PRICE : {predition.Price * 100000:C}");
            Console.WriteLine("--------------------------------------------------");

            Console.WriteLine("END PREDICT");

            Console.WriteLine("END MAIN FUNCTION");
        }

        #endregion
    }
}