Revize d358b79e

Re #8832 Label creation

Server/ServerApp/Predictor/NaiveBayesClassifier.cs

@@ -2 +2 @@
 // Author: Roman Kalivoda
 //
 
+using System;
 using System.Collections.Generic;
 using System.Linq;
 using Microsoft.ML;
-using ServerApp.Parser.OutputInfo;
 
 namespace ServerApp.Predictor
 {
@@ -17 +17 @@
         /// <summary>
         /// Context of the ML.NET framework.
         /// </summary>
-        private MLContext mlContext;
+        private MLContext _mlContext;
 
         /// <summary>
         /// Model instance
         /// </summary>
-        private ITransformer model;
+        private ITransformer _trainedModel;
+
+        private PredictionEngine<ModelInput, ModelOutput> _predictionEngine;
+
+        IDataView _trainingDataView;
 
         /// <summary>
         /// Instantiates new <c>MLContext</c>.
         /// </summary>
         public NaiveBayesClassifier()
         {
-            mlContext = new MLContext();
+            _mlContext = new MLContext();
         }
 
         public void Fit(IEnumerable<ModelInput> trainInput)
         {
-            IDataView trainingDataView = mlContext.Data.LoadFromEnumerable(trainInput);
-            var pipeline = mlContext.Transforms.Conversion.MapValueToKey(nameof(ModelInput.Label))
-                .Append(mlContext.Transforms.Concatenate("Features", new[] { "Temp" }))
-                .Append(mlContext.Transforms.NormalizeMinMax("Features", "Features"))
-                .Append(mlContext.MulticlassClassification.Trainers.NaiveBayes());
+            this._trainingDataView = _mlContext.Data.LoadFromEnumerable(trainInput);
+            var pipeline = _mlContext.Transforms.Conversion.MapValueToKey(nameof(ModelInput.Label))
+                .Append(_mlContext.Transforms.Concatenate("Features", new[] { "Temp" }))
+                .Append(_mlContext.Transforms.NormalizeMinMax("Features", "Features"))
+                .AppendCacheCheckpoint(_mlContext)
+                .Append(_mlContext.MulticlassClassification.Trainers.NaiveBayes())
+                .Append(_mlContext.Transforms.Conversion.MapKeyToValue("PredictedLabel")); ;
+
+            this._trainedModel = pipeline.Fit(this._trainingDataView);
+            this._predictionEngine = _mlContext.Model.CreatePredictionEngine<ModelInput, ModelOutput>(this._trainedModel);
 
-            this.model =pipeline.Fit(trainingDataView);
+        }
 
+        public string Predict(ModelInput input)
+        {
+            return this._predictionEngine.Predict(input).Prediction;
         }
 
-        public IDataView Predict(IEnumerable<ModelInput> input)
+        public void Evaluate(IEnumerable<ModelInput> modelInputs)
         {
-            var data = mlContext.Data.LoadFromEnumerable(input);
-            IDataView result = model.Transform(data);
-            return result;
+            var testDataView = this._mlContext.Data.LoadFromEnumerable(modelInputs);
+            var testMetrics = _mlContext.MulticlassClassification.Evaluate(_trainedModel.Transform(testDataView));
+
+            Console.WriteLine($"*************************************************************************************************************");
+            Console.WriteLine($"*       Metrics for Multi-class Classification model - Test Data     ");
+            Console.WriteLine($"*------------------------------------------------------------------------------------------------------------");
+            Console.WriteLine($"*       MicroAccuracy:    {testMetrics.MicroAccuracy:0.###}");
+            Console.WriteLine($"*       MacroAccuracy:    {testMetrics.MacroAccuracy:0.###}");
+            Console.WriteLine($"*       LogLoss:          {testMetrics.LogLoss:#.###}");
+            Console.WriteLine($"*       LogLossReduction: {testMetrics.LogLossReduction:#.###}");
+            Console.WriteLine($"*************************************************************************************************************");
         }
     }
 }