using Microsoft.ML;
using Microsoft.ML.Data;
using Microsoft.ML.Trainers;
using System;
using System.Collections.Generic;
using System.Linq;
using System.Reflection;
using System.Text;
using System.Threading.Tasks;
using System.Xml.Linq;

namespace testML
{
	internal class Program
	{
		static string[] tags = new string[] { "Rojo", "Amarillo", "Verde claro", "Verde oscuro", "Violeta", "Naranja", "Azul", "Blanco" };
		static Random rnd = new Random();

		static void Main(string[] args)
		{
			
			

			var tmpData = new List<Data>();
		

			for (var c = 0; c < 300; c++)
			{
				var d = CreateRandomData();
				tmpData.Add(d);
			}

			

			MLContext mlContext = new MLContext();

			var data = mlContext.Data.LoadFromEnumerable(tmpData);


			DataOperationsCatalog.TrainTestData dataSplit = mlContext.Data.TrainTestSplit(data, testFraction: 0.2);
			IDataView trainData = dataSplit.TrainSet;
			IDataView testData = dataSplit.TestSet;

			
			var trainer = mlContext.Regression.Trainers.Sdca(maximumNumberOfIterations:1000);
			//var trainer = mlContext.Regression.Trainers.OnlineGradientDescent(numberOfIterations: 100, learningRate: 0.01f );


			var pipeline = mlContext.Transforms.CopyColumns(outputColumnName: "Label", inputColumnName: "IntegerNumber")
				.Append(mlContext.Transforms.Text.NormalizeText("StringTest"))
				.Append(mlContext.Transforms.Text.FeaturizeText("StringTest"))
				.Append(mlContext.Transforms.Concatenate("Features", "Enum1", "Enum2", "Enum3", "Enum4", "StringTest"))
				.Append(mlContext.Transforms.NormalizeMinMax("Features"))
				.Append(trainer);


			
			ITransformer model = pipeline.Fit(trainData);
			

			// Use trained model to make inferences on test data
			IDataView testDataPredictions = model.Transform(testData);

			// Extract model metrics and get RSquared
			RegressionMetrics trainedModelMetrics = mlContext.Regression.Evaluate(testDataPredictions);
			double rSquared = trainedModelMetrics.RSquared;

			Console.WriteLine("ModelMetrics: {0}", rSquared);



			var predictionFunction = mlContext.Model.CreatePredictionEngine<Data, DataPrediction>(model);


			for (var c = 0; c < 1000; c++)
			{

				var test = CreateRandomData();
				var expected = test.IntegerNumber;
				test.IntegerNumber = 0;

				var p = predictionFunction.Predict(test);
				
				Console.WriteLine("Found: {0:#,##0.00}\tExpected: {1:#,##0.00}\t\tDiff: {2:#,##0.00}", p.IntegerNumber, expected , expected- p.IntegerNumber);
			}

		}

		private static Data CreateRandomData()
		{
			var d = new Data()
			{
				Accession = rnd.Next(0, 99999999).ToString("00000000"),
				Enum1 = rnd.Next(1, 4),
				Enum2 = rnd.Next(1, 11),
				Enum3 = rnd.Next(1, 6),
				Enum4 = rnd.Next(1, 6),
				StringTest = tags[rnd.Next(0, tags.Length)]
			};

			// Ponemos algunos datos que tengan alguna relación (la red neuronal debería calibrarse para comprender esta formula)
			d.IntegerNumber = (((d.Enum1 + d.Enum2) - (d.Enum3 + d.Enum4)) * 5.25f) + d.StringTest.Length;
			
			d.DecimalNumber = (d.Enum2 / d.Enum1) * (2.0f + (1.0f / d.StringTest.Length));

			if (d.StringTest == "Azul")
			{
				d.IntegerNumber += 10;
				d.OrigenResultNumber = 1;
			}

			if (d.StringTest == "Rojo")
			{
				d.IntegerNumber += 5f;
				d.OrigenResultNumber = 1;
			}

			return d;

		}
	}


}