Merge pull request #9 from Brainydaps/1.2.3

1.2.3

README.md

@@ -15,11 +15,17 @@ TalkingStage is a conversational bot that uses machine learning to predict respo
 ![Screenshot 2024-06-25 172912](https://github.com/Brainydaps/TalkingStage/assets/41041115/2a04c075-23b4-4ae5-a9c1-87972993eff6)
 ![Screenshot 2024-06-29 at 03 26 57](https://github.com/Brainydaps/TalkingStage/assets/41041115/b3a18c48-0370-4471-8e84-2e24a2abfdd3)
 
-## What's New in v1.2.1
+## What's New in v1.2.3 Release
+
+- Implemented a more advanced machine learning algorithm (LightGBM) for improved prediction accuracy.
+- Fine-tuned model parameters for enhanced performance and reliability.
+- Optimized data preprocessing steps to ensure consistent and effective training.
+- Updated data loading and preprocessing to handle text data more efficiently.
+- Streamlined the training pipeline for faster model training and deployment.
+- Enhanced model robustness with increased iterations and improved feature handling.
+- Integrated comprehensive error handling and logging for better debugging capabilities.
+- Improved overall code readability and maintainability.
 
-- **Initial Interaction Alert**: On the first interaction, the bot now displays an alert with guidance on how to format questions for better responses. This helps users understand the input format for optimal results.
-- **Unique Responses**: Updated the response mechanism to ensure that each response is unique and separated by a comma.
-- **Expanded Training Data**: The training data has been expanded to better handle a wider variety of questions, resulting in more accurate and relevant answers.
 
 ## Project Structure
 

TrainingModel.cs

@@ -4,6 +4,7 @@
 using System.Linq;
 using Microsoft.ML;
 using Microsoft.ML.Data;
+using Microsoft.ML.Trainers.LightGbm;
 using Microsoft.ML.Transforms.Text;
 
 namespace TalkingStage
@@ -34,17 +35,22 @@ public void TrainAndSaveModel(string trainingDataPath, string modelPath)
             var preprocessedTrainingDataView = mlContext.Data.LoadFromEnumerable(preprocessedTrainingData);
 
             // Define the data preparation and training pipeline
-            var pipeline = mlContext.Transforms.Text.FeaturizeText("Features", nameof(TalkingStageBot.InputData.Text))
-                .Append(mlContext.Transforms.Conversion.MapValueToKey("LabelKey", nameof(TalkingStageBot.InputData.Label)))
-                .Append(mlContext.Transforms.Concatenate("Features", "Features"))
-                .AppendCacheCheckpoint(mlContext)
-                .Append(mlContext.MulticlassClassification.Trainers.SdcaMaximumEntropy(
-                    labelColumnName: "LabelKey",
-                    featureColumnName: "Features",
-                    l2Regularization: 0.1f,
-                    l1Regularization: 0.01f,
-                    maximumNumberOfIterations: 1000))
-                .Append(mlContext.Transforms.Conversion.MapKeyToValue("PredictedLabel", "PredictedLabel"));
+            var pipeline = mlContext.Transforms.Text.FeaturizeText(inputColumnName: @"Text", outputColumnName: @"Text")
+                                    .Append(mlContext.Transforms.Concatenate(@"Features", new[] { @"Text" }))
+                                    .Append(mlContext.Transforms.Conversion.MapValueToKey(outputColumnName: @"Label", inputColumnName: @"Label", addKeyValueAnnotationsAsText: false))
+                                    .Append(mlContext.MulticlassClassification.Trainers.LightGbm(new LightGbmMulticlassTrainer.Options() { 
+                                        NumberOfLeaves = 30,
+                                        NumberOfIterations = 200, 
+                                        MinimumExampleCountPerLeaf = 1, 
+                                        LearningRate = 0.8, 
+                                        LabelColumnName = @"Label", FeatureColumnName = @"Features", ExampleWeightColumnName = null, 
+                                        Booster = new GradientBooster.Options() { 
+                                            SubsampleFraction = 1, 
+                                            FeatureFraction = 0.9, 
+                                            L1Regularization = 0.03, 
+                                            L2Regularization = 0.1 }, 
+                                        MaximumBinCountPerFeature = 254 }))
+                                    .Append(mlContext.Transforms.Conversion.MapKeyToValue(outputColumnName: @"PredictedLabel", inputColumnName: @"PredictedLabel"));
 
             // Train the model
             var model = pipeline.Fit(preprocessedTrainingDataView);