From 539a0f7b7afe3c465b3cea151c304b4005487e1f Mon Sep 17 00:00:00 2001
From: sierralee51 <36902786+sierralee51@users.noreply.github.com>
Date: Mon, 1 Jul 2019 13:12:44 -0700
Subject: [PATCH 01/44] reformatted multiclassclassification samples

---
 .../LbfgsMaximumEntropy.cs                    | 62 +++++++++-----
 .../LbfgsMaximumEntropy.tt                    |  3 +-
 .../LbfgsMaximumEntropyWithOptions.cs         | 71 ++++++++++------
 .../LbfgsMaximumEntropyWithOptions.tt         |  4 +-
 .../MulticlassClassification/LightGbm.cs      | 67 ++++++++++-----
 .../MulticlassClassification/LightGbm.tt      |  9 +-
 .../LightGbmWithOptions.cs                    | 82 ++++++++++++-------
 .../LightGbmWithOptions.tt                    |  8 +-
 .../LogLossPerClass.cs                        | 42 ++++++----
 .../MulticlassClassification.ttinclude        | 72 +++++++++++-----
 .../MulticlassClassification/NaiveBayes.cs    | 74 +++++++++++------
 .../MulticlassClassification/NaiveBayes.tt    | 13 +--
 .../MulticlassClassification/OneVersusAll.cs  | 62 +++++++++-----
 .../MulticlassClassification/OneVersusAll.tt  |  3 +-
 .../PairwiseCoupling.cs                       | 60 +++++++++-----
 .../PairwiseCoupling.tt                       |  4 +-
 .../PermutationFeatureImportance.cs           | 66 +++++++++------
 .../SdcaMaximumEntropy.cs                     | 72 ++++++++++------
 .../SdcaMaximumEntropy.tt                     |  3 +-
 .../SdcaMaximumEntropyWithOptions.cs          | 73 +++++++++++------
 .../SdcaMaximumEntropyWithOptions.tt          |  3 +-
 .../SdcaNonCalibrated.cs                      | 64 ++++++++++-----
 .../SdcaNonCalibrated.tt                      |  3 +-
 .../SdcaNonCalibratedWithOptions.cs           | 73 +++++++++++------
 .../SdcaNonCalibratedWithOptions.tt           | 27 +++---
 25 files changed, 668 insertions(+), 352 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index ab45d4ba53..f280564b7a 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -10,39 +10,50 @@ public static class LbfgsMaximumEntropy
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey(nameof(DataPoint.Label))
+                    mlContext.Transforms.Conversion
+                        .MapValueToKey(nameof(DataPoint.Label))
+
                     // Apply LbfgsMaximumEntropy multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.LbfgsMaximumEntropy());
+                    .Append(mlContext.MulticlassClassification.Trainers
+                        .LbfgsMaximumEntropy());
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " +
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -52,15 +63,17 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //  Micro Accuracy: 0.91
             //  Macro Accuracy: 0.91
             //  Log Loss: 0.24
             //  Log Loss Reduction: 0.79
-            
+
             //  Confusion table
             //            ||========================
             //  PREDICTED ||     0 |     1 |     2 | Recall
@@ -72,8 +85,11 @@ public static void Example()
             //  Precision ||0.9308 |0.9593 |0.8580 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -85,13 +101,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -114,8 +134,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.tt
index a58f4f0917..519c549bd9 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.tt
@@ -8,7 +8,8 @@ string TrainerOptions = null;
 string OptionsInclude = "";
 string Comments = "";
 bool CacheData = false;
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
+    + "\n        // range with labels 1, 2 or 3.";
 
 string ExpectedOutputPerInstance = @"// Expected output:
             //   Label: 1, Prediction: 1
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
index a818057677..5e0f57caab 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
@@ -11,48 +11,56 @@ public static class LbfgsMaximumEntropyWithOptions
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define trainer options.
             var options = new LbfgsMaximumEntropyMulticlassTrainer.Options
-                        {
-                            HistorySize = 50,
-                            L1Regularization = 0.1f,
-                            NumberOfThreads = 1
-                        };
+            {
+                HistorySize = 50,
+                L1Regularization = 0.1f,
+                NumberOfThreads = 1
+            };
 
             // Define the trainer.
-            var pipeline = 
-			        // Convert the string labels into key types.
+            var pipeline =
+                    // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply LbfgsMaximumEntropy multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.LbfgsMaximumEntropy(options));
-			
+                    .Append(mlContext.MulticlassClassification.Trainers
+                        .LbfgsMaximumEntropy(options));
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " +
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -62,9 +70,11 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
@@ -82,8 +92,11 @@ public static void Example()
             //   Precision ||0.9304 |0.9593 |0.8529 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -95,13 +108,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -124,8 +141,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.tt
index 12caedbd3b..1e96c1eab9 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.tt
@@ -12,7 +12,9 @@ string TrainerOptions = @"LbfgsMaximumEntropyMulticlassTrainer.Options
 
 string OptionsInclude = "using Microsoft.ML.Trainers;";
 string Comments = "";
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
+    + "\n        // range with labels 1, 2 or 3.";
+
 bool CacheData = false;
 
 string ExpectedOutputPerInstance = @"// Expected output:
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
index 84ac92ecef..3376bfcf33 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
@@ -8,43 +8,55 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 {
     public static class LightGbm
     {
-        // This example requires installation of additional NuGet package
-        // <a href="https://www.nuget.org/packages/Microsoft.ML.FastTree/">Microsoft.ML.FastTree</a>.
+        // This example requires installation of additional NuGet package for 
+        // Microsoft.ML.FastTree at
+        // https://www.nuget.org/packages/Microsoft.ML.FastTree/
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey(nameof(DataPoint.Label))
+                    mlContext.Transforms.Conversion
+                        .MapValueToKey(nameof(DataPoint.Label))
+
                     // Apply LightGbm multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.LightGbm());
+                    .Append(mlContext.MulticlassClassification.Trainers
+                        .LightGbm());
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " +
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -54,15 +66,17 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //   Micro Accuracy: 0.99
             //   Macro Accuracy: 0.99
             //   Log Loss: 0.05
             //   Log Loss Reduction: 0.95
-                 
+
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
@@ -74,8 +88,11 @@ public static void Example()
             //   Precision ||0.9936 |1.0000 |0.9701 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -87,13 +104,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -116,8 +137,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.tt
index 216ca2a0b6..47944467ba 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.tt
@@ -7,10 +7,13 @@ string TrainerOptions = null;
 
 string OptionsInclude = "";
 string Comments = @"
-        // This example requires installation of additional NuGet package
-        // <a href=""https://www.nuget.org/packages/Microsoft.ML.FastTree/"">Microsoft.ML.FastTree</a>.";
+        // This example requires installation of additional NuGet package for 
+        // Microsoft.ML.FastTree at
+        // https://www.nuget.org/packages/Microsoft.ML.FastTree/";
+
 bool CacheData = false;
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
+    + "\n        // range with labels 1, 2 or 3.";
 
 string ExpectedOutputPerInstance = @"// Expected output:
             //   Label: 1, Prediction: 1
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
index 3b4444081c..c5a7f42fef 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
@@ -9,54 +9,63 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 {
     public static class LightGbmWithOptions
     {
-        // This example requires installation of additional NuGet package
-        // <a href="https://www.nuget.org/packages/Microsoft.ML.FastTree/">Microsoft.ML.FastTree</a>.
+        // This example requires installation of additional NuGet package for 
+        // Microsoft.ML.FastTree at
+        // https://www.nuget.org/packages/Microsoft.ML.FastTree/
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define trainer options.
             var options = new LightGbmMulticlassTrainer.Options
-                        {
-                            Booster = new DartBooster.Options()
-                            {
-                                TreeDropFraction = 0.15,
-                                XgboostDartMode = false
-                            }
-                        };
+            {
+                Booster = new DartBooster.Options()
+                {
+                    TreeDropFraction = 0.15,
+                    XgboostDartMode = false
+                }
+            };
 
             // Define the trainer.
-            var pipeline = 
-			        // Convert the string labels into key types.
+            var pipeline =
+                    // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply LightGbm multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.LightGbm(options));
-			
+                    .Append(mlContext.MulticlassClassification.Trainers
+                        .LightGbm(options));
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " +
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -66,15 +75,17 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //   Micro Accuracy: 0.98
             //   Macro Accuracy: 0.98
             //   Log Loss: 0.07
             //   Log Loss Reduction: 0.94
-                 
+
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
@@ -86,8 +97,11 @@ public static void Example()
             //   Precision ||0.9936 |1.0000 |0.9419 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -99,13 +113,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -128,8 +146,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.tt
index 7ec6706227..b267db33f0 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.tt
@@ -14,11 +14,13 @@ string TrainerOptions = @"LightGbmMulticlassTrainer.Options
 
 string OptionsInclude = "using Microsoft.ML.Trainers.LightGbm;";
 string Comments = @"
-        // This example requires installation of additional NuGet package
-        // <a href=""https://www.nuget.org/packages/Microsoft.ML.FastTree/"">Microsoft.ML.FastTree</a>.";
+        // This example requires installation of additional NuGet package for 
+        // Microsoft.ML.FastTree at
+        // https://www.nuget.org/packages/Microsoft.ML.FastTree/";
 
 bool CacheData = false;
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
+    + "\n        // range with labels 1, 2 or 3.";
 
 string ExpectedOutputPerInstance = @"// Expected output:
             //   Label: 1, Prediction: 1
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs
index 86ab646083..c875616ada 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs
@@ -10,42 +10,50 @@ public static class LogLossPerClass
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey(nameof(DataPoint.Label))
+                    mlContext.Transforms.Conversion
+                        .MapValueToKey(nameof(DataPoint.Label))
+
                     // Apply a multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers.LightGbm());
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
 
             // Find the original label values.
             VBuffer<uint> keys = default;
             transformedTestData.Schema["PredictedLabel"].GetKeyValues(ref keys);
             var originalLabels = keys.DenseValues().ToArray();
             for (var i = 0; i < originalLabels.Length; i++)
-                Console.WriteLine($"LogLoss for label {originalLabels[i]}: {metrics.PerClassLogLoss[i]:F4}");
+                Console.WriteLine($"LogLoss for label " +
+                    $"{originalLabels[i]}: {metrics.PerClassLogLoss[i]:F4}");
 
             // Expected output:
             //   LogLoss for label 7: 0.2578
@@ -60,7 +68,9 @@ public static void Example()
         }
 
         // Generates data points with random features and labels 1 to 9.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed = 0)
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -72,13 +82,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -95,4 +109,4 @@ private class Prediction
             public uint PredictedLabel { get; set; }
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
index 87201f12a3..38b5a86dcf 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
@@ -13,21 +13,25 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
     {<#=Comments#>
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 <# if (CacheData) { #>
 
-            // ML.NET doesn't cache data set by default. Therefore, if one reads a data set from a file and accesses it many times,
-			// it can be slow due to expensive featurization and disk operations. When the considered data can fit into memory,
-			// a solution is to cache the data in memory. Caching is especially helpful when working with iterative algorithms 
+            // ML.NET doesn't cache data set by default. Therefore, if one reads a
+			// data set from a file and accesses it many times, it can be slow due
+			// to expensive featurization and disk operations. When the considered
+			// data can fit into memory, a solution is to cache the data in memory.
+			// Caching is especially helpful when working with iterative algorithms 
 			// which needs many data passes.
 			trainingData = mlContext.Data.Cache(trainingData);
 <# } #>
@@ -37,15 +41,23 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply <#=MetaTrainer#> multiclass meta trainer on top of binary trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.<#=MetaTrainer#>(<#=Trainer#>()));
+                    // Apply <#=MetaTrainer#> multiclass meta trainer on top of
+					// binary trainer.
+                    .Append(mlContext.MulticlassClassification.Trainers
+					    .<#=MetaTrainer#>(
+						<#=Trainer#>()));
+
 <# } else if (TrainerOptions == null) { #>
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey(nameof(DataPoint.Label))
+                    mlContext.Transforms.Conversion
+					    .MapValueToKey(nameof(DataPoint.Label))
+
                     // Apply <#=Trainer#> multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.<#=Trainer#>());
+                    .Append(mlContext.MulticlassClassification.Trainers
+					    .<#=Trainer#>());
+
 <# } else { #>
             // Define trainer options.
             var options = new <#=TrainerOptions#>;
@@ -55,37 +67,47 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 			        // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply <#=Trainer#> multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.<#=Trainer#>(options));
+                    .Append(mlContext.MulticlassClassification.Trainers
+					    .<#=Trainer#>(options));
 			
 <# } #>
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+			// from training data.
+            var testData = mlContext.Data
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             <#=ExpectedOutputPerInstance#>
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+			    .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
             
             <#=ExpectedOutput#>
         }
 
         <#=DataGenerationComments#>
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+		    int seed=0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -97,13 +119,17 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+					// constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -126,7 +152,9 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
index d76006bf2c..aa4c714873 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
@@ -9,44 +9,57 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
     public static class NaiveBayes
     {
         // Naive Bayes classifier is based on Bayes' theorem. 
-        // It assumes independence among the presence of features in a class even though they may be dependent on each other.
-        // It is  a multi-class trainer that accepts binary feature values of type float, i.e., feature values are either true or false.
-        // Specifically a feature value greater than zero is treated as true, zero or less is treated as false.
+        // It assumes independence among the presence of features in a class even
+        // though they may be dependent on each other. It is  a multi-class trainer
+        // that accepts binary feature values of type float, i.e., feature values
+        // are either true or false. Specifically a feature value greater than zero
+        // is treated as true, zero or less is treated as false.
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey(nameof(DataPoint.Label))
+                    mlContext.Transforms.Conversion
+                        .MapValueToKey(nameof(DataPoint.Label))
+
                     // Apply NaiveBayes multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.NaiveBayes());
+                    .Append(mlContext.MulticlassClassification.Trainers
+                        .NaiveBayes());
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " +
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -56,15 +69,17 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //   Micro Accuracy: 0.88
             //   Macro Accuracy: 0.88
             //   Log Loss: 34.54
             //   Log Loss Reduction: -30.47
-                 
+
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
@@ -76,10 +91,13 @@ public static void Example()
             //   Precision ||0.9467 |0.8735 |0.8061 |
         }
 
-        
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        // For NaiveBayes values greater than zero are treated as true, zero or less are treated as false.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+
+        // Generates random uniform doubles in [-0.5, 0.5) range with labels
+        // 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
+        // zero or less are treated as false.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -91,13 +109,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -120,8 +142,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
index a74ad64a80..04ace790a7 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
@@ -9,13 +9,16 @@ bool CacheData = false;
 string OptionsInclude = "";
 string Comments= @"
         // Naive Bayes classifier is based on Bayes' theorem. 
-        // It assumes independence among the presence of features in a class even though they may be dependent on each other.
-        // It is  a multi-class trainer that accepts binary feature values of type float, i.e., feature values are either true or false.
-        // Specifically a feature value greater than zero is treated as true, zero or less is treated as false.";
+        // It assumes independence among the presence of features in a class even
+		// though they may be dependent on each other. It is  a multi-class trainer
+		// that accepts binary feature values of type float, i.e., feature values
+		// are either true or false. Specifically a feature value greater than zero
+		// is treated as true, zero or less is treated as false.";
 
 string DataGenerationComments= @"
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        // For NaiveBayes values greater than zero are treated as true, zero or less are treated as false.";
+        // Generates random uniform doubles in [-0.5, 0.5) range with labels
+		// 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
+		// zero or less are treated as false.";
 
 string ExpectedOutputPerInstance= @"// Expected output:
             //   Label: 1, Prediction: 1
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
index 9b93fd3cd0..3893736257 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
@@ -10,39 +10,50 @@ public static class OneVersusAll
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply OneVersusAll multiclass meta trainer on top of binary trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.OneVersusAll(mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+                    // Apply OneVersusAll multiclass meta trainer on top of
+                    // binary trainer.
+                    .Append(mlContext.MulticlassClassification.Trainers
+                        .OneVersusAll(
+                        mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " +
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -52,15 +63,17 @@ public static void Example()
             //   Label: 3, Prediction: 2
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //   Micro Accuracy: 0.90
             //   Macro Accuracy: 0.90
             //   Log Loss: 0.36
             //   Log Loss Reduction: 0.68
-            
+
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
@@ -72,8 +85,11 @@ public static void Example()
             //   Precision ||0.8994 |0.9180 |0.8851 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -85,13 +101,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -114,8 +134,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.tt
index 0233d9948e..ce355c56b0 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.tt
@@ -7,7 +7,8 @@ string TrainerOptions = null;
 
 string OptionsInclude = "";
 string Comments= "";
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
+    + "\n        // range with labels 1, 2 or 3.";
 
 bool CacheData = false;
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
index de0dab8a9c..6b00af2c39 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
@@ -10,39 +10,50 @@ public static class PairwiseCoupling
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply PairwiseCoupling multiclass meta trainer on top of binary trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.PairwiseCoupling(mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+                    // Apply PairwiseCoupling multiclass meta trainer on top of
+                    // binary trainer.
+                    .Append(mlContext.MulticlassClassification.Trainers
+                        .PairwiseCoupling(
+                        mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " +
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -52,9 +63,11 @@ public static void Example()
             //   Label: 3, Prediction: 2
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //   Micro Accuracy: 0.90
             //   Macro Accuracy: 0.90
@@ -72,8 +85,11 @@ public static void Example()
             //   Precision ||0.9091 |0.9171 |0.8636 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -85,13 +101,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -114,8 +134,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.tt
index 4f3617e693..a568fd9736 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.tt
@@ -7,7 +7,9 @@ string TrainerOptions = null;
 
 string OptionsInclude = "";
 string Comments= "";
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
+    + "\n        // range with labels 1, 2 or 3.";
+
 bool CacheData = false;
 
 string ExpectedOutputPerInstance= @"// Expected output:
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PermutationFeatureImportance.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PermutationFeatureImportance.cs
index 963fd238ca..714c69361b 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PermutationFeatureImportance.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PermutationFeatureImportance.cs
@@ -9,9 +9,10 @@ public static class PermutationFeatureImportance
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            var mlContext = new MLContext(seed:1);
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness.
+            var mlContext = new MLContext(seed: 1);
 
             // Create sample data.
             var samples = GenerateData();
@@ -19,13 +20,17 @@ public static void Example()
             // Load the sample data as an IDataView.
             var data = mlContext.Data.LoadFromEnumerable(samples);
 
-            // Define a training pipeline that concatenates features into a vector, normalizes them, and then
-            // trains a linear model.
-            var featureColumns = new string[] { nameof(Data.Feature1), nameof(Data.Feature2) };
-            var pipeline = mlContext.Transforms.Concatenate("Features", featureColumns)
+            // Define a training pipeline that concatenates features into a vector,
+            // normalizes them, and then trains a linear model.
+            var featureColumns =
+                new string[] { nameof(Data.Feature1), nameof(Data.Feature2) };
+
+            var pipeline = mlContext.Transforms
+                    .Concatenate("Features", featureColumns)
                     .Append(mlContext.Transforms.Conversion.MapValueToKey("Label"))
                     .Append(mlContext.Transforms.NormalizeMinMax("Features"))
-                    .Append(mlContext.MulticlassClassification.Trainers.SdcaMaximumEntropy());
+                    .Append(mlContext.MulticlassClassification.Trainers
+                    .SdcaMaximumEntropy());
 
             // Fit the pipeline to the data.
             var model = pipeline.Fit(data);
@@ -36,18 +41,26 @@ public static void Example()
             // Extract the predictor.
             var linearPredictor = model.LastTransformer;
 
-            // Compute the permutation metrics for the linear model using the normalized data.
-            var permutationMetrics = mlContext.MulticlassClassification.PermutationFeatureImportance(
-                linearPredictor, transformedData, permutationCount: 30);
-
-            // Now let's look at which features are most important to the model overall.
-            // Get the feature indices sorted by their impact on microaccuracy.
-            var sortedIndices = permutationMetrics.Select((metrics, index) => new { index, metrics.MicroAccuracy})
+            // Compute the permutation metrics for the linear model using the
+            // normalized data.
+            var permutationMetrics = mlContext.MulticlassClassification
+                .PermutationFeatureImportance(linearPredictor, transformedData,
+                permutationCount: 30);
+
+            // Now let's look at which features are most important to the model
+            // overall. Get the feature indices sorted by their impact on
+            // microaccuracy.
+            var sortedIndices = permutationMetrics
+                .Select((metrics, index) => new { index, metrics.MicroAccuracy })
                 .OrderByDescending(feature => Math.Abs(feature.MicroAccuracy.Mean))
                 .Select(feature => feature.index);
 
-            Console.WriteLine("Feature\tChange in MicroAccuracy\t95% Confidence in the Mean Change in MicroAccuracy");
-            var microAccuracy = permutationMetrics.Select(x => x.MicroAccuracy).ToArray();
+            Console.WriteLine("Feature\tChange in MicroAccuracy\t95% Confidence in "
+                + "the Mean Change in MicroAccuracy");
+
+            var microAccuracy = permutationMetrics.Select(x => x.MicroAccuracy)
+                .ToArray();
+
             foreach (int i in sortedIndices)
             {
                 Console.WriteLine("{0}\t{1:G4}\t{2:G4}",
@@ -76,10 +89,14 @@ private class Data
         /// linear combination of the features.
         /// </summary>
         /// <param name="nExamples">The number of examples.</param>
-        /// <param name="bias">The bias, or offset, in the calculation of the label.</param>
-        /// <param name="weight1">The weight to multiply the first feature with to compute the label.</param>
-        /// <param name="weight2">The weight to multiply the second feature with to compute the label.</param>
-        /// <param name="seed">The seed for generating feature values and label noise.</param>
+        /// <param name="bias">The bias, or offset, in the calculation of the
+        /// label.</param>
+        /// <param name="weight1">The weight to multiply the first feature with to
+        /// compute the label.</param>
+        /// <param name="weight2">The weight to multiply the second feature with to
+        /// compute the label.</param>
+        /// <param name="seed">The seed for generating feature values and label
+        /// noise.</param>
         /// <returns>An enumerable of Data objects.</returns>
         private static IEnumerable<Data> GenerateData(int nExamples = 10000,
             double bias = 0, double weight1 = 1, double weight2 = 2, int seed = 1)
@@ -95,7 +112,10 @@ private static IEnumerable<Data> GenerateData(int nExamples = 10000,
                 };
 
                 // Create a noisy label.
-                var value = (float)(bias + weight1 * data.Feature1 + weight2 * data.Feature2 + rng.NextDouble() - 0.5);
+                var value = (float)
+                    (bias + weight1 * data.Feature1 + weight2 * data.Feature2 +
+                    rng.NextDouble() - 0.5);
+
                 if (value < max / 3)
                     data.Label = 0;
                 else if (value < 2 * max / 3)
@@ -106,4 +126,4 @@ private static IEnumerable<Data> GenerateData(int nExamples = 10000,
             }
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
index 1cc66a33cc..966cc14155 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
@@ -10,45 +10,58 @@ public static class SdcaMaximumEntropy
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
-            // ML.NET doesn't cache data set by default. Therefore, if one reads a data set from a file and accesses it many times,
-			// it can be slow due to expensive featurization and disk operations. When the considered data can fit into memory,
-			// a solution is to cache the data in memory. Caching is especially helpful when working with iterative algorithms 
-			// which needs many data passes.
-			trainingData = mlContext.Data.Cache(trainingData);
+            // ML.NET doesn't cache data set by default. Therefore, if one reads a
+            // data set from a file and accesses it many times, it can be slow due
+            // to expensive featurization and disk operations. When the considered
+            // data can fit into memory, a solution is to cache the data in memory.
+            // Caching is especially helpful when working with iterative algorithms 
+            // which needs many data passes.
+            trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey(nameof(DataPoint.Label))
+                    mlContext.Transforms.Conversion
+                        .MapValueToKey(nameof(DataPoint.Label))
+
                     // Apply SdcaMaximumEntropy multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.SdcaMaximumEntropy());
+                    .Append(mlContext.MulticlassClassification.Trainers
+                        .SdcaMaximumEntropy());
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " +
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -58,9 +71,11 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
@@ -77,8 +92,11 @@ public static void Example()
             //   Precision ||0.9130 |0.9538 |0.8494 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -90,13 +108,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -119,8 +141,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.tt
index 4eccf95005..f83d618268 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.tt
@@ -8,7 +8,8 @@ string TrainerOptions = null;
 string OptionsInclude = "";
 string Comments = "";
 bool CacheData = true;
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)" 
+    + "\n        // range with labels 1, 2 or 3.";
 
 string ExpectedOutputPerInstance = @"// Expected output:
             //   Label: 1, Prediction: 1
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
index a3cfcf4cbc..d4205cb71d 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
@@ -11,22 +11,26 @@ public static class SdcaMaximumEntropyWithOptions
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
-            // ML.NET doesn't cache data set by default. Therefore, if one reads a data set from a file and accesses it many times,
-			// it can be slow due to expensive featurization and disk operations. When the considered data can fit into memory,
-			// a solution is to cache the data in memory. Caching is especially helpful when working with iterative algorithms 
-			// which needs many data passes.
-			trainingData = mlContext.Data.Cache(trainingData);
+            // ML.NET doesn't cache data set by default. Therefore, if one reads a
+            // data set from a file and accesses it many times, it can be slow due
+            // to expensive featurization and disk operations. When the considered
+            // data can fit into memory, a solution is to cache the data in memory.
+            // Caching is especially helpful when working with iterative algorithms 
+            // which needs many data passes.
+            trainingData = mlContext.Data.Cache(trainingData);
 
             // Define trainer options.
             var options = new SdcaMaximumEntropyMulticlassTrainer.Options
@@ -38,28 +42,34 @@ public static void Example()
             };
 
             // Define the trainer.
-            var pipeline = 
-			        // Convert the string labels into key types.
+            var pipeline =
+                    // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply SdcaMaximumEntropy multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.SdcaMaximumEntropy(options));
-			
+                    .Append(mlContext.MulticlassClassification.Trainers
+                        .SdcaMaximumEntropy(options));
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " +
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -69,9 +79,11 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //   Micro Accuracy: 0.92
             //   Macro Accuracy: 0.92
@@ -89,8 +101,11 @@ public static void Example()
             //   Precision ||0.9363 |0.9647 |0.8497 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -102,13 +117,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -131,8 +150,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.tt
index 88eead7e0d..f6bb41ac40 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.tt
@@ -14,7 +14,8 @@ string TrainerOptions = @"SdcaMaximumEntropyMulticlassTrainer.Options
 string OptionsInclude = "using Microsoft.ML.Trainers;";
 string Comments = "";
 bool CacheData = true;
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)" 
+    + "\n        // range with labels 1, 2 or 3.";
 
 
 string ExpectedOutputPerInstance = @"// Expected output:
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index d9f59fe3e7..912e409e81 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -10,45 +10,58 @@ public static class SdcaNonCalibrated
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
-            // ML.NET doesn't cache data set by default. Therefore, if one reads a data set from a file and accesses it many times,
-			// it can be slow due to expensive featurization and disk operations. When the considered data can fit into memory,
-			// a solution is to cache the data in memory. Caching is especially helpful when working with iterative algorithms 
+            // ML.NET doesn't cache data set by default. Therefore, if one reads a
+			// data set from a file and accesses it many times, it can be slow due
+			// to expensive featurization and disk operations. When the considered
+			// data can fit into memory, a solution is to cache the data in memory.
+			// Caching is especially helpful when working with iterative algorithms 
 			// which needs many data passes.
 			trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey(nameof(DataPoint.Label))
+                    mlContext.Transforms.Conversion
+					    .MapValueToKey(nameof(DataPoint.Label))
+
                     // Apply SdcaNonCalibrated multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.SdcaNonCalibrated());
+                    .Append(mlContext.MulticlassClassification.Trainers
+					    .SdcaNonCalibrated());
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+			// from training data.
+            var testData = mlContext.Data
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -58,7 +71,9 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+			    .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
             
             // Expected output:
@@ -78,8 +93,11 @@ public static void Example()
             //   Precision ||0.9304 |0.9538 |0.8521 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+		    int seed=0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -91,13 +109,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+					// constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -120,7 +142,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt
index 5437228267..cd403b1a44 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt
@@ -8,7 +8,8 @@ string TrainerOptions = null;
 string OptionsInclude = "";
 string Comments = "";
 bool CacheData = true;
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
+    + "\n        // range with labels 1, 2 or 3.";
 
 string ExpectedOutputPerInstance = @"// Expected output:
             //   Label: 1, Prediction: 1
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
index 2577756d9b..9f1cd73e9b 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
@@ -11,22 +11,26 @@ public static class SdcaNonCalibratedWithOptions
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
-            // ML.NET doesn't cache data set by default. Therefore, if one reads a data set from a file and accesses it many times,
-			// it can be slow due to expensive featurization and disk operations. When the considered data can fit into memory,
-			// a solution is to cache the data in memory. Caching is especially helpful when working with iterative algorithms 
-			// which needs many data passes.
-			trainingData = mlContext.Data.Cache(trainingData);
+            // ML.NET doesn't cache data set by default. Therefore, if one reads a
+            // data set from a file and accesses it many times, it can be slow due
+            // to expensive featurization and disk operations. When the considered
+            // data can fit into memory, a solution is to cache the data in memory.
+            // Caching is especially helpful when working with iterative algorithms 
+            // which needs many data passes.
+            trainingData = mlContext.Data.Cache(trainingData);
 
             // Define trainer options.
             var options = new SdcaNonCalibratedMulticlassTrainer.Options
@@ -38,28 +42,34 @@ public static void Example()
             };
 
             // Define the trainer.
-            var pipeline = 
-			        // Convert the string labels into key types.
+            var pipeline =
+                    // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply SdcaNonCalibrated multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.SdcaNonCalibrated(options));
-			
+                    .Append(mlContext.MulticlassClassification.Trainers
+                        .SdcaNonCalibrated(options));
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+            // from training data.
+            var testData = mlContext.Data
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " +
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -69,9 +79,11 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
+            var metrics = mlContext.MulticlassClassification
+                .Evaluate(transformedTestData);
+
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
@@ -89,8 +101,11 @@ public static void Example()
             //   Precision ||0.9236 |0.9591 |0.8372 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed = 0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -102,13 +117,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+                    // constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -131,8 +150,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.tt
index 19f7ee9dd9..cd403b1a44 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.tt
@@ -1,20 +1,15 @@
 <#@ include file="MulticlassClassification.ttinclude"#>
 <#+
-string ClassName = "SdcaNonCalibratedWithOptions";
+string ClassName = "SdcaNonCalibrated";
 string Trainer = "SdcaNonCalibrated";
 string MetaTrainer = null;
-string TrainerOptions = @"SdcaNonCalibratedMulticlassTrainer.Options
-            {
-                Loss = new HingeLoss(),
-                L1Regularization = 0.1f,
-                BiasLearningRate = 0.01f,
-                NumberOfThreads = 1
-            }";
+string TrainerOptions = null;
 
-string OptionsInclude = "using Microsoft.ML.Trainers;";
+string OptionsInclude = "";
 string Comments = "";
 bool CacheData = true;
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
+    + "\n        // range with labels 1, 2 or 3.";
 
 string ExpectedOutputPerInstance = @"// Expected output:
             //   Label: 1, Prediction: 1
@@ -26,16 +21,16 @@ string ExpectedOutputPerInstance = @"// Expected output:
 string ExpectedOutput = @"// Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
-            //   Log Loss: 0.22
-            //   Log Loss Reduction: 0.80
+            //   Log Loss: 0.57
+            //   Log Loss Reduction: 0.48
 
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
             //   TRUTH     ||========================
-            //           0 ||   145 |     0 |    15 | 0.9063
-            //           1 ||     0 |   164 |    13 | 0.9266
-            //           2 ||    12 |     7 |   144 | 0.8834
+            //           0 ||   147 |     0 |    13 | 0.9188
+            //           1 ||     0 |   165 |    12 | 0.9322
+            //           2 ||    11 |     8 |   144 | 0.8834
             //             ||========================
-            //   Precision ||0.9236 |0.9591 |0.8372 |";
+            //   Precision ||0.9304 |0.9538 |0.8521 |";
 #>
\ No newline at end of file

From d18c419fef235cce982f92f4c0788959ce001d3b Mon Sep 17 00:00:00 2001
From: sierralee51 <36902786+sierralee51@users.noreply.github.com>
Date: Mon, 1 Jul 2019 13:17:47 -0700
Subject: [PATCH 02/44] fixing errors

---
 .../SdcaMaximumEntropyWithOptions.cs          | 56 ++++++-------
 .../SdcaNonCalibrated.cs                      | 72 ++++++-----------
 .../SdcaNonCalibrated.tt                      |  3 +-
 .../SdcaNonCalibratedWithOptions.cs           | 78 +++++++++----------
 4 files changed, 88 insertions(+), 121 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
index d4205cb71d..ab030b46b2 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
@@ -12,25 +12,25 @@ public static class SdcaMaximumEntropyWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-            // exception tracking and logging, as a catalog of available operations
-            // and as the source of randomness. Setting the seed to a fixed number
-            // in this example to make outputs deterministic.
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-            // consumable by ML.NET API.
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-            // data set from a file and accesses it many times, it can be slow due
-            // to expensive featurization and disk operations. When the considered
-            // data can fit into memory, a solution is to cache the data in memory.
-            // Caching is especially helpful when working with iterative algorithms 
-            // which needs many data passes.
-            trainingData = mlContext.Data.Cache(trainingData);
+			// data set from a file and accesses it many times, it can be slow due
+			// to expensive featurization and disk operations. When the considered
+			// data can fit into memory, a solution is to cache the data in memory.
+			// Caching is especially helpful when working with iterative algorithms 
+			// which needs many data passes.
+			trainingData = mlContext.Data.Cache(trainingData);
 
             // Define trainer options.
             var options = new SdcaMaximumEntropyMulticlassTrainer.Options
@@ -42,34 +42,34 @@ public static void Example()
             };
 
             // Define the trainer.
-            var pipeline =
-                    // Convert the string labels into key types.
+            var pipeline = 
+			        // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply SdcaMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .SdcaMaximumEntropy(options));
-
+					    .SdcaMaximumEntropy(options));
+			
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-            // from training data.
+			// from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-                reuseRowObject: false).ToList();
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " +
-                    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -80,10 +80,10 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-                .Evaluate(transformedTestData);
+			    .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
-
+            
             // Expected output:
             //   Micro Accuracy: 0.92
             //   Macro Accuracy: 0.92
@@ -104,7 +104,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-            int seed = 0)
+		    int seed=0)
 
         {
             var random = new Random(seed);
@@ -118,16 +118,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-                    // constant multiple of label.
+					// constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -151,9 +151,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index 912e409e81..b63449958c 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -10,58 +10,45 @@ public static class SdcaNonCalibrated
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
+            // as a catalog of available operations and as the source of randomness.
+            // Setting the seed to a fixed number in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
-            // ML.NET doesn't cache data set by default. Therefore, if one reads a
-			// data set from a file and accesses it many times, it can be slow due
-			// to expensive featurization and disk operations. When the considered
-			// data can fit into memory, a solution is to cache the data in memory.
-			// Caching is especially helpful when working with iterative algorithms 
-			// which needs many data passes.
-			trainingData = mlContext.Data.Cache(trainingData);
+            // ML.NET doesn't cache data set by default. Therefore, if one reads a data set from a file and accesses it many times,
+            // it can be slow due to expensive featurization and disk operations. When the considered data can fit into memory,
+            // a solution is to cache the data in memory. Caching is especially helpful when working with iterative algorithms 
+            // which needs many data passes.
+            trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion
-					    .MapValueToKey(nameof(DataPoint.Label))
-
+                    mlContext.Transforms.Conversion.MapValueToKey(nameof(DataPoint.Label))
                     // Apply SdcaNonCalibrated multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
-					    .SdcaNonCalibrated());
-
+                    .Append(mlContext.MulticlassClassification.Trainers.SdcaNonCalibrated());
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different
-			// from training data.
-            var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different from training data.
+            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -71,11 +58,9 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
-
+            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
             PrintMetrics(metrics);
-            
+
             // Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
@@ -93,11 +78,8 @@ public static void Example()
             //   Precision ||0.9304 |0.9538 |0.8521 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5)
-        // range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
-
+        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed = 0)
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -109,17 +91,13 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a
-					// constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
-
+                    // The feature values are slightly increased by adding a constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+        // Example with label and 20 feature values. A data set is a collection of such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -142,10 +120,8 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
-
+            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
+}
\ No newline at end of file
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt
index cd403b1a44..5437228267 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt
@@ -8,8 +8,7 @@ string TrainerOptions = null;
 string OptionsInclude = "";
 string Comments = "";
 bool CacheData = true;
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
-    + "\n        // range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
 
 string ExpectedOutputPerInstance = @"// Expected output:
             //   Label: 1, Prediction: 1
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
index 9f1cd73e9b..912e409e81 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
@@ -3,73 +3,65 @@
 using System.Linq;
 using Microsoft.ML;
 using Microsoft.ML.Data;
-using Microsoft.ML.Trainers;
 
 namespace Samples.Dynamic.Trainers.MulticlassClassification
 {
-    public static class SdcaNonCalibratedWithOptions
+    public static class SdcaNonCalibrated
     {
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-            // exception tracking and logging, as a catalog of available operations
-            // and as the source of randomness. Setting the seed to a fixed number
-            // in this example to make outputs deterministic.
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-            // consumable by ML.NET API.
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-            // data set from a file and accesses it many times, it can be slow due
-            // to expensive featurization and disk operations. When the considered
-            // data can fit into memory, a solution is to cache the data in memory.
-            // Caching is especially helpful when working with iterative algorithms 
-            // which needs many data passes.
-            trainingData = mlContext.Data.Cache(trainingData);
-
-            // Define trainer options.
-            var options = new SdcaNonCalibratedMulticlassTrainer.Options
-            {
-                Loss = new HingeLoss(),
-                L1Regularization = 0.1f,
-                BiasLearningRate = 0.01f,
-                NumberOfThreads = 1
-            };
+			// data set from a file and accesses it many times, it can be slow due
+			// to expensive featurization and disk operations. When the considered
+			// data can fit into memory, a solution is to cache the data in memory.
+			// Caching is especially helpful when working with iterative algorithms 
+			// which needs many data passes.
+			trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey("Label")
+                    mlContext.Transforms.Conversion
+					    .MapValueToKey(nameof(DataPoint.Label))
+
                     // Apply SdcaNonCalibrated multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .SdcaNonCalibrated(options));
+					    .SdcaNonCalibrated());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-            // from training data.
+			// from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-                reuseRowObject: false).ToList();
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " +
-                    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -80,31 +72,31 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-                .Evaluate(transformedTestData);
+			    .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
-
+            
             // Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
-            //   Log Loss: 0.22
-            //   Log Loss Reduction: 0.80
+            //   Log Loss: 0.57
+            //   Log Loss Reduction: 0.48
 
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
             //   TRUTH     ||========================
-            //           0 ||   145 |     0 |    15 | 0.9063
-            //           1 ||     0 |   164 |    13 | 0.9266
-            //           2 ||    12 |     7 |   144 | 0.8834
+            //           0 ||   147 |     0 |    13 | 0.9188
+            //           1 ||     0 |   165 |    12 | 0.9322
+            //           2 ||    11 |     8 |   144 | 0.8834
             //             ||========================
-            //   Precision ||0.9236 |0.9591 |0.8372 |
+            //   Precision ||0.9304 |0.9538 |0.8521 |
         }
 
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-            int seed = 0)
+		    int seed=0)
 
         {
             var random = new Random(seed);
@@ -118,16 +110,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-                    // constant multiple of label.
+					// constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -151,9 +143,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}

From 0b4aff68f031246785aadce04e9f6d39d509434d Mon Sep 17 00:00:00 2001
From: sierralee51 <36902786+sierralee51@users.noreply.github.com>
Date: Mon, 1 Jul 2019 13:36:26 -0700
Subject: [PATCH 03/44] reformatted MulticlassClassification samples

---
 .../LbfgsMaximumEntropy.cs                    | 42 +++++------
 .../LbfgsMaximumEntropyWithOptions.cs         | 54 +++++++-------
 .../MulticlassClassification/LightGbm.cs      | 42 +++++------
 .../LightGbmWithOptions.cs                    | 60 ++++++++--------
 .../MulticlassClassification/NaiveBayes.cs    | 56 +++++++--------
 .../MulticlassClassification/OneVersusAll.cs  | 44 ++++++------
 .../PairwiseCoupling.cs                       | 42 +++++------
 .../SdcaMaximumEntropy.cs                     | 52 +++++++-------
 .../SdcaNonCalibrated.cs                      | 72 ++++++++++++-------
 .../SdcaNonCalibrated.tt                      |  3 +-
 .../SdcaNonCalibratedWithOptions.cs           | 36 ++++++----
 .../SdcaNonCalibratedWithOptions.tt           | 24 ++++---
 12 files changed, 283 insertions(+), 244 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index f280564b7a..59adefec38 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -11,49 +11,49 @@ public static class LbfgsMaximumEntropy
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-            // exception tracking and logging, as a catalog of available operations
-            // and as the source of randomness. Setting the seed to a fixed number
-            // in this example to make outputs deterministic.
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-            // consumable by ML.NET API.
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-                        .MapValueToKey(nameof(DataPoint.Label))
+					    .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply LbfgsMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .LbfgsMaximumEntropy());
+					    .LbfgsMaximumEntropy());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-            // from training data.
+			// from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-                reuseRowObject: false).ToList();
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " +
-                    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -64,16 +64,16 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-                .Evaluate(transformedTestData);
+			    .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
-
+            
             // Expected output:
             //  Micro Accuracy: 0.91
             //  Macro Accuracy: 0.91
             //  Log Loss: 0.24
             //  Log Loss Reduction: 0.79
-
+            
             //  Confusion table
             //            ||========================
             //  PREDICTED ||     0 |     1 |     2 | Recall
@@ -88,7 +88,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-            int seed = 0)
+		    int seed=0)
 
         {
             var random = new Random(seed);
@@ -102,16 +102,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-                    // constant multiple of label.
+					// constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -135,9 +135,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
index 5e0f57caab..4f33bdeb2b 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
@@ -12,55 +12,55 @@ public static class LbfgsMaximumEntropyWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-            // exception tracking and logging, as a catalog of available operations
-            // and as the source of randomness. Setting the seed to a fixed number
-            // in this example to make outputs deterministic.
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-            // consumable by ML.NET API.
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define trainer options.
             var options = new LbfgsMaximumEntropyMulticlassTrainer.Options
-            {
-                HistorySize = 50,
-                L1Regularization = 0.1f,
-                NumberOfThreads = 1
-            };
+                        {
+                            HistorySize = 50,
+                            L1Regularization = 0.1f,
+                            NumberOfThreads = 1
+                        };
 
             // Define the trainer.
-            var pipeline =
-                    // Convert the string labels into key types.
+            var pipeline = 
+			        // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply LbfgsMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .LbfgsMaximumEntropy(options));
-
+					    .LbfgsMaximumEntropy(options));
+			
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-            // from training data.
+			// from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-                reuseRowObject: false).ToList();
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " +
-                    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -71,10 +71,10 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-                .Evaluate(transformedTestData);
+			    .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
-
+            
             // Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
@@ -95,7 +95,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-            int seed = 0)
+		    int seed=0)
 
         {
             var random = new Random(seed);
@@ -109,16 +109,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-                    // constant multiple of label.
+					// constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -142,9 +142,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
index 3376bfcf33..2092a7785b 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
@@ -14,49 +14,49 @@ public static class LightGbm
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-            // exception tracking and logging, as a catalog of available operations
-            // and as the source of randomness. Setting the seed to a fixed number
-            // in this example to make outputs deterministic.
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-            // consumable by ML.NET API.
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-                        .MapValueToKey(nameof(DataPoint.Label))
+					    .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply LightGbm multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .LightGbm());
+					    .LightGbm());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-            // from training data.
+			// from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-                reuseRowObject: false).ToList();
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " +
-                    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -67,16 +67,16 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-                .Evaluate(transformedTestData);
+			    .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
-
+            
             // Expected output:
             //   Micro Accuracy: 0.99
             //   Macro Accuracy: 0.99
             //   Log Loss: 0.05
             //   Log Loss Reduction: 0.95
-
+                 
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
@@ -91,7 +91,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-            int seed = 0)
+		    int seed=0)
 
         {
             var random = new Random(seed);
@@ -105,16 +105,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-                    // constant multiple of label.
+					// constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -138,9 +138,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
index c5a7f42fef..c68a2cd032 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
@@ -15,57 +15,57 @@ public static class LightGbmWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-            // exception tracking and logging, as a catalog of available operations
-            // and as the source of randomness. Setting the seed to a fixed number
-            // in this example to make outputs deterministic.
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-            // consumable by ML.NET API.
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define trainer options.
             var options = new LightGbmMulticlassTrainer.Options
-            {
-                Booster = new DartBooster.Options()
-                {
-                    TreeDropFraction = 0.15,
-                    XgboostDartMode = false
-                }
-            };
+                        {
+                            Booster = new DartBooster.Options()
+                            {
+                                TreeDropFraction = 0.15,
+                                XgboostDartMode = false
+                            }
+                        };
 
             // Define the trainer.
-            var pipeline =
-                    // Convert the string labels into key types.
+            var pipeline = 
+			        // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply LightGbm multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .LightGbm(options));
-
+					    .LightGbm(options));
+			
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-            // from training data.
+			// from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-                reuseRowObject: false).ToList();
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " +
-                    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -76,16 +76,16 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-                .Evaluate(transformedTestData);
+			    .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
-
+            
             // Expected output:
             //   Micro Accuracy: 0.98
             //   Macro Accuracy: 0.98
             //   Log Loss: 0.07
             //   Log Loss Reduction: 0.94
-
+                 
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
@@ -100,7 +100,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-            int seed = 0)
+		    int seed=0)
 
         {
             var random = new Random(seed);
@@ -114,16 +114,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-                    // constant multiple of label.
+					// constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -147,9 +147,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
index aa4c714873..1e654c25fd 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
@@ -10,56 +10,56 @@ public static class NaiveBayes
     {
         // Naive Bayes classifier is based on Bayes' theorem. 
         // It assumes independence among the presence of features in a class even
-        // though they may be dependent on each other. It is  a multi-class trainer
-        // that accepts binary feature values of type float, i.e., feature values
-        // are either true or false. Specifically a feature value greater than zero
-        // is treated as true, zero or less is treated as false.
+		// though they may be dependent on each other. It is  a multi-class trainer
+		// that accepts binary feature values of type float, i.e., feature values
+		// are either true or false. Specifically a feature value greater than zero
+		// is treated as true, zero or less is treated as false.
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-            // exception tracking and logging, as a catalog of available operations
-            // and as the source of randomness. Setting the seed to a fixed number
-            // in this example to make outputs deterministic.
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-            // consumable by ML.NET API.
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-                        .MapValueToKey(nameof(DataPoint.Label))
+					    .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply NaiveBayes multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .NaiveBayes());
+					    .NaiveBayes());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-            // from training data.
+			// from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-                reuseRowObject: false).ToList();
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " +
-                    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -70,16 +70,16 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-                .Evaluate(transformedTestData);
+			    .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
-
+            
             // Expected output:
             //   Micro Accuracy: 0.88
             //   Macro Accuracy: 0.88
             //   Log Loss: 34.54
             //   Log Loss Reduction: -30.47
-
+                 
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
@@ -91,12 +91,12 @@ public static void Example()
             //   Precision ||0.9467 |0.8735 |0.8061 |
         }
 
-
+        
         // Generates random uniform doubles in [-0.5, 0.5) range with labels
-        // 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
-        // zero or less are treated as false.
+		// 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
+		// zero or less are treated as false.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-            int seed = 0)
+		    int seed=0)
 
         {
             var random = new Random(seed);
@@ -110,16 +110,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-                    // constant multiple of label.
+					// constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -143,9 +143,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
index 3893736257..001641831d 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
@@ -11,16 +11,16 @@ public static class OneVersusAll
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-            // exception tracking and logging, as a catalog of available operations
-            // and as the source of randomness. Setting the seed to a fixed number
-            // in this example to make outputs deterministic.
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-            // consumable by ML.NET API.
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
@@ -28,32 +28,32 @@ public static void Example()
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply OneVersusAll multiclass meta trainer on top of
-                    // binary trainer.
+					// binary trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .OneVersusAll(
-                        mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+					    .OneVersusAll(
+						mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-            // from training data.
+			// from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-                reuseRowObject: false).ToList();
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " +
-                    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -64,16 +64,16 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-                .Evaluate(transformedTestData);
+			    .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
-
+            
             // Expected output:
             //   Micro Accuracy: 0.90
             //   Macro Accuracy: 0.90
             //   Log Loss: 0.36
             //   Log Loss Reduction: 0.68
-
+            
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
@@ -88,7 +88,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-            int seed = 0)
+		    int seed=0)
 
         {
             var random = new Random(seed);
@@ -102,16 +102,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-                    // constant multiple of label.
+					// constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -135,9 +135,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
index 6b00af2c39..e9f49f4454 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
@@ -11,16 +11,16 @@ public static class PairwiseCoupling
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-            // exception tracking and logging, as a catalog of available operations
-            // and as the source of randomness. Setting the seed to a fixed number
-            // in this example to make outputs deterministic.
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-            // consumable by ML.NET API.
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
@@ -28,32 +28,32 @@ public static void Example()
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply PairwiseCoupling multiclass meta trainer on top of
-                    // binary trainer.
+					// binary trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .PairwiseCoupling(
-                        mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+					    .PairwiseCoupling(
+						mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-            // from training data.
+			// from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-                reuseRowObject: false).ToList();
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " +
-                    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -64,10 +64,10 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-                .Evaluate(transformedTestData);
+			    .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
-
+            
             // Expected output:
             //   Micro Accuracy: 0.90
             //   Macro Accuracy: 0.90
@@ -88,7 +88,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-            int seed = 0)
+		    int seed=0)
 
         {
             var random = new Random(seed);
@@ -102,16 +102,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-                    // constant multiple of label.
+					// constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -135,9 +135,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
index 966cc14155..82832b06bd 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
@@ -11,57 +11,57 @@ public static class SdcaMaximumEntropy
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-            // exception tracking and logging, as a catalog of available operations
-            // and as the source of randomness. Setting the seed to a fixed number
-            // in this example to make outputs deterministic.
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-            // consumable by ML.NET API.
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-            // data set from a file and accesses it many times, it can be slow due
-            // to expensive featurization and disk operations. When the considered
-            // data can fit into memory, a solution is to cache the data in memory.
-            // Caching is especially helpful when working with iterative algorithms 
-            // which needs many data passes.
-            trainingData = mlContext.Data.Cache(trainingData);
+			// data set from a file and accesses it many times, it can be slow due
+			// to expensive featurization and disk operations. When the considered
+			// data can fit into memory, a solution is to cache the data in memory.
+			// Caching is especially helpful when working with iterative algorithms 
+			// which needs many data passes.
+			trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-                        .MapValueToKey(nameof(DataPoint.Label))
+					    .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply SdcaMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .SdcaMaximumEntropy());
+					    .SdcaMaximumEntropy());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-            // from training data.
+			// from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-                reuseRowObject: false).ToList();
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, " +
-                    $"Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -72,10 +72,10 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-                .Evaluate(transformedTestData);
+			    .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
-
+            
             // Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
@@ -95,7 +95,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-            int seed = 0)
+		    int seed=0)
 
         {
             var random = new Random(seed);
@@ -109,16 +109,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-                    // constant multiple of label.
+					// constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                        .Select(x => randomFloat() + label * 0.2f).ToArray()
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -142,9 +142,9 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index b63449958c..912e409e81 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -10,45 +10,58 @@ public static class SdcaNonCalibrated
     {
         public static void Example()
         {
-            // Create a new context for ML.NET operations. It can be used for exception tracking and logging, 
-            // as a catalog of available operations and as the source of randomness.
-            // Setting the seed to a fixed number in this example to make outputs deterministic.
+            // Create a new context for ML.NET operations. It can be used for
+			// exception tracking and logging, as a catalog of available operations
+			// and as the source of randomness. Setting the seed to a fixed number
+			// in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
-            // Convert the list of data points to an IDataView object, which is consumable by ML.NET API.
+            // Convert the list of data points to an IDataView object, which is
+			// consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
-            // ML.NET doesn't cache data set by default. Therefore, if one reads a data set from a file and accesses it many times,
-            // it can be slow due to expensive featurization and disk operations. When the considered data can fit into memory,
-            // a solution is to cache the data in memory. Caching is especially helpful when working with iterative algorithms 
-            // which needs many data passes.
-            trainingData = mlContext.Data.Cache(trainingData);
+            // ML.NET doesn't cache data set by default. Therefore, if one reads a
+			// data set from a file and accesses it many times, it can be slow due
+			// to expensive featurization and disk operations. When the considered
+			// data can fit into memory, a solution is to cache the data in memory.
+			// Caching is especially helpful when working with iterative algorithms 
+			// which needs many data passes.
+			trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey(nameof(DataPoint.Label))
+                    mlContext.Transforms.Conversion
+					    .MapValueToKey(nameof(DataPoint.Label))
+
                     // Apply SdcaNonCalibrated multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.SdcaNonCalibrated());
+                    .Append(mlContext.MulticlassClassification.Trainers
+					    .SdcaNonCalibrated());
+
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
-            // Create testing data. Use different random seed to make it different from training data.
-            var testData = mlContext.Data.LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            // Create testing data. Use different random seed to make it different
+			// from training data.
+            var testData = mlContext.Data
+			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
-            var predictions = mlContext.Data.CreateEnumerable<Prediction>(transformedTestData, reuseRowObject: false).ToList();
+            var predictions = mlContext.Data
+			    .CreateEnumerable<Prediction>(transformedTestData,
+				reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
-                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+                Console.WriteLine($"Label: {p.Label}, " + 
+				    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -58,9 +71,11 @@ public static void Example()
             //   Label: 3, Prediction: 3
 
             // Evaluate the overall metrics
-            var metrics = mlContext.MulticlassClassification.Evaluate(transformedTestData);
-            PrintMetrics(metrics);
+            var metrics = mlContext.MulticlassClassification
+			    .Evaluate(transformedTestData);
 
+            PrintMetrics(metrics);
+            
             // Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
@@ -78,8 +93,11 @@ public static void Example()
             //   Precision ||0.9304 |0.9538 |0.8521 |
         }
 
-        // Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed = 0)
+        // Generates random uniform doubles in [-0.5, 0.5)
+        // range with labels 1, 2 or 3.
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+		    int seed=0)
+
         {
             var random = new Random(seed);
             float randomFloat() => (float)(random.NextDouble() - 0.5);
@@ -91,13 +109,17 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int se
                 {
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
-                    // The feature values are slightly increased by adding a constant multiple of label.
-                    Features = Enumerable.Repeat(label, 20).Select(x => randomFloat() + label * 0.2f).ToArray()
+                    // The feature values are slightly increased by adding a
+					// constant multiple of label.
+                    Features = Enumerable.Repeat(label, 20)
+					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+
                 };
             }
         }
 
-        // Example with label and 20 feature values. A data set is a collection of such examples.
+        // Example with label and 20 feature values. A data set is a collection of
+		// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -120,8 +142,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Micro Accuracy: {metrics.MicroAccuracy:F2}");
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
-            Console.WriteLine($"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            Console.WriteLine(
+			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt
index 5437228267..cd403b1a44 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.tt
@@ -8,7 +8,8 @@ string TrainerOptions = null;
 string OptionsInclude = "";
 string Comments = "";
 bool CacheData = true;
-string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5) range with labels 1, 2 or 3.";
+string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
+    + "\n        // range with labels 1, 2 or 3.";
 
 string ExpectedOutputPerInstance = @"// Expected output:
             //   Label: 1, Prediction: 1
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
index 912e409e81..da44e924aa 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
@@ -3,10 +3,11 @@
 using System.Linq;
 using Microsoft.ML;
 using Microsoft.ML.Data;
+using Microsoft.ML.Trainers;
 
 namespace Samples.Dynamic.Trainers.MulticlassClassification
 {
-    public static class SdcaNonCalibrated
+    public static class SdcaNonCalibratedWithOptions
     {
         public static void Example()
         {
@@ -31,16 +32,23 @@ public static void Example()
 			// which needs many data passes.
 			trainingData = mlContext.Data.Cache(trainingData);
 
-            // Define the trainer.
-            var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion
-					    .MapValueToKey(nameof(DataPoint.Label))
+            // Define trainer options.
+            var options = new SdcaNonCalibratedMulticlassTrainer.Options
+            {
+                Loss = new HingeLoss(),
+                L1Regularization = 0.1f,
+                BiasLearningRate = 0.01f,
+                NumberOfThreads = 1
+            };
 
+            // Define the trainer.
+            var pipeline = 
+			        // Convert the string labels into key types.
+                    mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply SdcaNonCalibrated multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .SdcaNonCalibrated());
-
+					    .SdcaNonCalibrated(options));
+			
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
@@ -79,18 +87,18 @@ public static void Example()
             // Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
-            //   Log Loss: 0.57
-            //   Log Loss Reduction: 0.48
+            //   Log Loss: 0.22
+            //   Log Loss Reduction: 0.80
 
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
             //   TRUTH     ||========================
-            //           0 ||   147 |     0 |    13 | 0.9188
-            //           1 ||     0 |   165 |    12 | 0.9322
-            //           2 ||    11 |     8 |   144 | 0.8834
+            //           0 ||   145 |     0 |    15 | 0.9063
+            //           1 ||     0 |   164 |    13 | 0.9266
+            //           2 ||    12 |     7 |   144 | 0.8834
             //             ||========================
-            //   Precision ||0.9304 |0.9538 |0.8521 |
+            //   Precision ||0.9236 |0.9591 |0.8372 |
         }
 
         // Generates random uniform doubles in [-0.5, 0.5)
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.tt
index cd403b1a44..5f0f2a4775 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.tt
@@ -1,11 +1,17 @@
 <#@ include file="MulticlassClassification.ttinclude"#>
 <#+
-string ClassName = "SdcaNonCalibrated";
+string ClassName = "SdcaNonCalibratedWithOptions";
 string Trainer = "SdcaNonCalibrated";
 string MetaTrainer = null;
-string TrainerOptions = null;
+string TrainerOptions = @"SdcaNonCalibratedMulticlassTrainer.Options
+            {
+                Loss = new HingeLoss(),
+                L1Regularization = 0.1f,
+                BiasLearningRate = 0.01f,
+                NumberOfThreads = 1
+            }";
 
-string OptionsInclude = "";
+string OptionsInclude = "using Microsoft.ML.Trainers;";
 string Comments = "";
 bool CacheData = true;
 string DataGenerationComments= "// Generates random uniform doubles in [-0.5, 0.5)"
@@ -21,16 +27,16 @@ string ExpectedOutputPerInstance = @"// Expected output:
 string ExpectedOutput = @"// Expected output:
             //   Micro Accuracy: 0.91
             //   Macro Accuracy: 0.91
-            //   Log Loss: 0.57
-            //   Log Loss Reduction: 0.48
+            //   Log Loss: 0.22
+            //   Log Loss Reduction: 0.80
 
             //   Confusion table
             //             ||========================
             //   PREDICTED ||     0 |     1 |     2 | Recall
             //   TRUTH     ||========================
-            //           0 ||   147 |     0 |    13 | 0.9188
-            //           1 ||     0 |   165 |    12 | 0.9322
-            //           2 ||    11 |     8 |   144 | 0.8834
+            //           0 ||   145 |     0 |    15 | 0.9063
+            //           1 ||     0 |   164 |    13 | 0.9266
+            //           2 ||    12 |     7 |   144 | 0.8834
             //             ||========================
-            //   Precision ||0.9304 |0.9538 |0.8521 |";
+            //   Precision ||0.9236 |0.9591 |0.8372 |";
 #>
\ No newline at end of file

From f84939f07c215aa9540735232a2725bda2075a8c Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:43:16 -0700
Subject: [PATCH 04/44] Update LbfgsMaximumEntropy.cs

getting rid of whitespace
---
 .../MulticlassClassification/LbfgsMaximumEntropy.cs       | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index 59adefec38..538fd79635 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -11,16 +11,16 @@ public static class LbfgsMaximumEntropy
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.

From 130c6daab193c01c0d4ee0178bae01bb1cc89321 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:43:41 -0700
Subject: [PATCH 05/44] Update LbfgsMaximumEntropy.cs

---
 .../Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs    | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index 538fd79635..ca9ee99501 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -38,7 +38,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 

From 71e4c198295832a422404a44af5a772d1c714abc Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:44:46 -0700
Subject: [PATCH 06/44] Update LbfgsMaximumEntropyWithOptions.cs

getting rid of whitespace
---
 .../LbfgsMaximumEntropyWithOptions.cs                | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
index 4f33bdeb2b..8e9153e42b 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
@@ -12,16 +12,16 @@ public static class LbfgsMaximumEntropyWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define trainer options.
@@ -45,7 +45,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -118,7 +118,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From 8166627c354d06433c5215f827a41b224ef4894c Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:45:49 -0700
Subject: [PATCH 07/44] Update LightGbmWithOptions.cs

fixing whitespace
---
 .../MulticlassClassification/LightGbmWithOptions.cs  | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
index c68a2cd032..824abf418a 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
@@ -15,16 +15,16 @@ public static class LightGbmWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define trainer options.
@@ -50,7 +50,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -123,7 +123,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From 5ea92f3f10be159b78c87f470291f222661d2cc0 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:46:26 -0700
Subject: [PATCH 08/44] Update LbfgsMaximumEntropy.cs

---
 .../Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs    | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index ca9ee99501..af2b3ab2f5 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -111,7 +111,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From b63c891061297b8f2112f43bff561a107cb99334 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:47:25 -0700
Subject: [PATCH 09/44] Update LightGbm.cs

fixing whitespace
---
 .../Trainers/MulticlassClassification/LightGbm.cs    | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
index 2092a7785b..1c39dad324 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
@@ -14,16 +14,16 @@ public static class LightGbm
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
@@ -41,7 +41,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -114,7 +114,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+	// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From 4d37328acf94b647ae04eca25645b92211453a8f Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:50:52 -0700
Subject: [PATCH 10/44] Update LightGbm.cs

fixing whitespace
---
 .../Dynamic/Trainers/MulticlassClassification/LightGbm.cs       | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
index 1c39dad324..34df10ea8b 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
@@ -105,7 +105,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
 					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 

From 6421a8baf53deb63e9c7aad2d90d700dc182fa05 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:52:09 -0700
Subject: [PATCH 11/44] Update LightGbmWithOptions.cs

fixing whitespace
---
 .../Trainers/MulticlassClassification/LightGbmWithOptions.cs  | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
index 824abf418a..16c73ff68b 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
@@ -39,7 +39,7 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-			        // Convert the string labels into key types.
+	            // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply LightGbm multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
@@ -114,7 +114,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
 					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 

From 79093c83543a9b447ca19b87a3d92601fddf4281 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:53:46 -0700
Subject: [PATCH 12/44] Update MulticlassClassification.ttinclude

fixing whitespace
---
 .../MulticlassClassification.ttinclude        | 20 +++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
index 38b5a86dcf..2efcd99a97 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
@@ -14,16 +14,16 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 <# if (CacheData) { #>
 
@@ -42,7 +42,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply <#=MetaTrainer#> multiclass meta trainer on top of
-					// binary trainer.
+		    // binary trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
 					    .<#=MetaTrainer#>(
 						<#=Trainer#>()));
@@ -64,7 +64,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 
             // Define the trainer.
             var pipeline = 
-			        // Convert the string labels into key types.
+                    // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply <#=Trainer#> multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
@@ -76,7 +76,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -120,7 +120,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
 					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -129,7 +129,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+	// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -158,4 +158,4 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
-}
\ No newline at end of file
+}

From 4c2f7235d29adf3f5a13b98fc5c0bc1c88949475 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:54:42 -0700
Subject: [PATCH 13/44] Update MulticlassClassification.ttinclude

fixing whitespace
---
 .../MulticlassClassification.ttinclude                 | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
index 2efcd99a97..9790c6a359 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
@@ -28,11 +28,11 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 <# if (CacheData) { #>
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-			// data set from a file and accesses it many times, it can be slow due
-			// to expensive featurization and disk operations. When the considered
-			// data can fit into memory, a solution is to cache the data in memory.
-			// Caching is especially helpful when working with iterative algorithms 
-			// which needs many data passes.
+	    // data set from a file and accesses it many times, it can be slow due
+	    // to expensive featurization and disk operations. When the considered
+	    // data can fit into memory, a solution is to cache the data in memory.
+	    // Caching is especially helpful when working with iterative algorithms 
+	    // which needs many data passes.
 			trainingData = mlContext.Data.Cache(trainingData);
 <# } #>
 

From 4a48dc5db416cadfaf422ba0c78a66ae627f5071 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:56:04 -0700
Subject: [PATCH 14/44] Update NaiveBayes.cs

fixing whitespace
---
 .../MulticlassClassification/NaiveBayes.cs    | 26 +++++++++----------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
index 1e654c25fd..354352b3c7 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
@@ -10,23 +10,23 @@ public static class NaiveBayes
     {
         // Naive Bayes classifier is based on Bayes' theorem. 
         // It assumes independence among the presence of features in a class even
-		// though they may be dependent on each other. It is  a multi-class trainer
-		// that accepts binary feature values of type float, i.e., feature values
-		// are either true or false. Specifically a feature value greater than zero
-		// is treated as true, zero or less is treated as false.
+	// though they may be dependent on each other. It is  a multi-class trainer
+	// that accepts binary feature values of type float, i.e., feature values
+	// are either true or false. Specifically a feature value greater than zero
+	// is treated as true, zero or less is treated as false.
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
@@ -44,7 +44,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -93,8 +93,8 @@ public static void Example()
 
         
         // Generates random uniform doubles in [-0.5, 0.5) range with labels
-		// 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
-		// zero or less are treated as false.
+	// 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
+	// zero or less are treated as false.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
 		    int seed=0)
 
@@ -110,7 +110,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
 					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -119,7 +119,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+	// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From cc1b3f8c79bd6ef630941071333ac24d03f60bb5 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:56:38 -0700
Subject: [PATCH 15/44] Update NaiveBayes.tt

fixing whitespace
---
 .../MulticlassClassification/NaiveBayes.tt         | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
index 04ace790a7..ca83a2b378 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
@@ -10,15 +10,15 @@ string OptionsInclude = "";
 string Comments= @"
         // Naive Bayes classifier is based on Bayes' theorem. 
         // It assumes independence among the presence of features in a class even
-		// though they may be dependent on each other. It is  a multi-class trainer
-		// that accepts binary feature values of type float, i.e., feature values
-		// are either true or false. Specifically a feature value greater than zero
-		// is treated as true, zero or less is treated as false.";
+        // though they may be dependent on each other. It is  a multi-class trainer
+	// that accepts binary feature values of type float, i.e., feature values
+	// are either true or false. Specifically a feature value greater than zero
+	// is treated as true, zero or less is treated as false.";
 
 string DataGenerationComments= @"
         // Generates random uniform doubles in [-0.5, 0.5) range with labels
-		// 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
-		// zero or less are treated as false.";
+	// 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
+	// zero or less are treated as false.";
 
 string ExpectedOutputPerInstance= @"// Expected output:
             //   Label: 1, Prediction: 1
@@ -42,4 +42,4 @@ string ExpectedOutput = @"// Expected output:
             //           2 ||     9 |    21 |   133 | 0.8160
             //             ||========================
             //   Precision ||0.9467 |0.8735 |0.8061 |";
-#>
\ No newline at end of file
+#>

From c6d4cfd7bb3cdcc606180293c432ac54fd5813d7 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:57:09 -0700
Subject: [PATCH 16/44] Update NaiveBayes.tt


From 97b3c9959404868c56d1ca00200dbfb6acf2885c Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:58:07 -0700
Subject: [PATCH 17/44] Update OneVersusAll.cs

fixing whitespace
---
 .../MulticlassClassification/OneVersusAll.cs     | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
index 001641831d..bab768a9d7 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
@@ -11,16 +11,16 @@ public static class OneVersusAll
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
@@ -28,7 +28,7 @@ public static void Example()
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply OneVersusAll multiclass meta trainer on top of
-					// binary trainer.
+		    // binary trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
 					    .OneVersusAll(
 						mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
@@ -38,7 +38,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -102,7 +102,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
 					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -111,7 +111,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From 6be7d8ae57733338777db46582f525311fac1bf2 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 13:59:01 -0700
Subject: [PATCH 18/44] Update PairwiseCoupling.cs

fixing whitespace
---
 .../MulticlassClassification/PairwiseCoupling.cs   | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
index e9f49f4454..f933989392 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
@@ -11,16 +11,16 @@ public static class PairwiseCoupling
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
@@ -38,7 +38,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -102,7 +102,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
 					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -111,7 +111,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From 3f7418790670b8c1aa80cebf7d66fc301e9f3004 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:00:26 -0700
Subject: [PATCH 19/44] Update SdcaMaximumEntropy.cs

fixing whitespace
---
 .../SdcaMaximumEntropy.cs                     | 24 +++++++++----------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
index 82832b06bd..be4cf22296 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
@@ -11,24 +11,24 @@ public static class SdcaMaximumEntropy
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-			// data set from a file and accesses it many times, it can be slow due
-			// to expensive featurization and disk operations. When the considered
-			// data can fit into memory, a solution is to cache the data in memory.
-			// Caching is especially helpful when working with iterative algorithms 
-			// which needs many data passes.
+	    // data set from a file and accesses it many times, it can be slow due
+	    // to expensive featurization and disk operations. When the considered
+	    // data can fit into memory, a solution is to cache the data in memory.
+	    // Caching is especially helpful when working with iterative algorithms 
+	    // which needs many data passes.
 			trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
@@ -46,7 +46,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -109,7 +109,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
 					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -118,7 +118,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+	// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From 0d23ad83845f9c6232f3da7754aaeb90a8bf5032 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:02:59 -0700
Subject: [PATCH 20/44] Update SdcaMaximumEntropyWithOptions.cs

fixing whitespace
---
 .../SdcaMaximumEntropyWithOptions.cs          | 26 +++++++++----------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
index ab030b46b2..0501abb458 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
@@ -12,24 +12,24 @@ public static class SdcaMaximumEntropyWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-			// data set from a file and accesses it many times, it can be slow due
-			// to expensive featurization and disk operations. When the considered
-			// data can fit into memory, a solution is to cache the data in memory.
-			// Caching is especially helpful when working with iterative algorithms 
-			// which needs many data passes.
+	    // data set from a file and accesses it many times, it can be slow due
+	    // to expensive featurization and disk operations. When the considered
+	    // data can fit into memory, a solution is to cache the data in memory.
+	    // Caching is especially helpful when working with iterative algorithms 
+	    // which needs many data passes.
 			trainingData = mlContext.Data.Cache(trainingData);
 
             // Define trainer options.
@@ -43,7 +43,7 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-			        // Convert the string labels into key types.
+	            // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply SdcaMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
@@ -54,7 +54,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -118,7 +118,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
 					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -127,7 +127,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+	// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From c728577cd70a6149d0843a0eef802fe30b0601d2 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:04:17 -0700
Subject: [PATCH 21/44] Update SdcaNonCalibrated.cs

fixing whitespace
---
 .../SdcaNonCalibrated.cs                      | 24 +++++++++----------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index 912e409e81..cf3d4d5f09 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -11,24 +11,24 @@ public static class SdcaNonCalibrated
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-			// data set from a file and accesses it many times, it can be slow due
-			// to expensive featurization and disk operations. When the considered
-			// data can fit into memory, a solution is to cache the data in memory.
-			// Caching is especially helpful when working with iterative algorithms 
-			// which needs many data passes.
+	    // data set from a file and accesses it many times, it can be slow due
+	    // to expensive featurization and disk operations. When the considered
+	    // data can fit into memory, a solution is to cache the data in memory.
+	    // Caching is especially helpful when working with iterative algorithms 
+	    // which needs many data passes.
 			trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
@@ -46,7 +46,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -110,7 +110,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
 					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -119,7 +119,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From 6e82a100af43b975d5152afda0276f9e131b54b7 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:05:48 -0700
Subject: [PATCH 22/44] Update SdcaNonCalibratedWithOptions.cs

fixing whitespace
---
 .../SdcaNonCalibratedWithOptions.cs           | 28 +++++++++----------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
index da44e924aa..44e40ff6df 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
@@ -12,25 +12,25 @@ public static class SdcaNonCalibratedWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-			// exception tracking and logging, as a catalog of available operations
-			// and as the source of randomness. Setting the seed to a fixed number
-			// in this example to make outputs deterministic.
+	    // exception tracking and logging, as a catalog of available operations
+	    // and as the source of randomness. Setting the seed to a fixed number
+	    // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-			// consumable by ML.NET API.
+	    // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-			// data set from a file and accesses it many times, it can be slow due
-			// to expensive featurization and disk operations. When the considered
-			// data can fit into memory, a solution is to cache the data in memory.
-			// Caching is especially helpful when working with iterative algorithms 
-			// which needs many data passes.
-			trainingData = mlContext.Data.Cache(trainingData);
+	    // data set from a file and accesses it many times, it can be slow due
+	    // to expensive featurization and disk operations. When the considered
+	    // data can fit into memory, a solution is to cache the data in memory.
+	    // Caching is especially helpful when working with iterative algorithms 
+	    // which needs many data passes.
+	    trainingData = mlContext.Data.Cache(trainingData);
 
             // Define trainer options.
             var options = new SdcaNonCalibratedMulticlassTrainer.Options
@@ -43,7 +43,7 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-			        // Convert the string labels into key types.
+	            // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply SdcaNonCalibrated multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
@@ -54,7 +54,7 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-			// from training data.
+	    // from training data.
             var testData = mlContext.Data
 			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -118,7 +118,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
 					    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -127,7 +127,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-		// such examples.
+	// such examples.
         private class DataPoint
         {
             public uint Label { get; set; }

From 1f1ff3b4aa2437a54e1b5e63aec1a08ad5ed4f61 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:07:04 -0700
Subject: [PATCH 23/44] Update SdcaNonCalibrated.cs

fixing whitespace
---
 .../Trainers/MulticlassClassification/SdcaNonCalibrated.cs      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index cf3d4d5f09..a128c7a65f 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -29,7 +29,7 @@ public static void Example()
 	    // data can fit into memory, a solution is to cache the data in memory.
 	    // Caching is especially helpful when working with iterative algorithms 
 	    // which needs many data passes.
-			trainingData = mlContext.Data.Cache(trainingData);
+	    trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =

From 239e26ad33fb0cef8386954b94233778ede210ee Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:08:05 -0700
Subject: [PATCH 24/44] Update SdcaNonCalibrated.cs

---
 .../Trainers/MulticlassClassification/SdcaNonCalibrated.cs      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index a128c7a65f..cf33d4ae1b 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -56,7 +56,7 @@ public static void Example()
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
 			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+			    reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))

From 32db971a935efa1cb7acda5ab33ec846fac6f860 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:15:41 -0700
Subject: [PATCH 25/44] Update LbfgsMaximumEntropy.cs

---
 .../LbfgsMaximumEntropy.cs                    | 22 +++++++++----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index af2b3ab2f5..db93fac36e 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -27,11 +27,11 @@ public static void Example()
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-					    .MapValueToKey(nameof(DataPoint.Label))
+		        .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply LbfgsMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .LbfgsMaximumEntropy());
+	                .LbfgsMaximumEntropy());
 
 
             // Train the model.
@@ -40,20 +40,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+	        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+	            $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -64,7 +64,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -87,8 +87,8 @@ public static void Example()
 
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, 
+	    int seed=0)
 
         {
             var random = new Random(seed);
@@ -102,9 +102,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+	                .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }

From e3e595bebd4512a2b716e8df75f4d839885ae264 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:17:08 -0700
Subject: [PATCH 26/44] Update LbfgsMaximumEntropy.cs

---
 .../Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs    | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index db93fac36e..d24869e6a4 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -135,7 +135,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From 969af4df73502a84fb22a1422c9b18898387ff0c Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:19:09 -0700
Subject: [PATCH 27/44] Update LbfgsMaximumEntropyWithOptions.cs

---
 .../LbfgsMaximumEntropyWithOptions.cs         | 22 +++++++++----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
index 8e9153e42b..89728c59f9 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
@@ -34,11 +34,11 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-			        // Convert the string labels into key types.
+	            // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply LbfgsMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .LbfgsMaximumEntropy(options));
+	                .LbfgsMaximumEntropy(options));
 			
 
             // Train the model.
@@ -47,20 +47,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+	        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -71,7 +71,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -95,7 +95,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -109,9 +109,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-					// constant multiple of label.
+		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+		        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -142,7 +142,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From b18c50853a5dfd45eb73fc2dfc212e9ba5b6e391 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:22:33 -0700
Subject: [PATCH 28/44] Update LightGbm.cs

---
 .../MulticlassClassification/LightGbm.cs      | 20 +++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
index 34df10ea8b..0d898b6790 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
@@ -30,11 +30,11 @@ public static void Example()
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-					    .MapValueToKey(nameof(DataPoint.Label))
+		        .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply LightGbm multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .LightGbm());
+			.LightGbm());
 
 
             // Train the model.
@@ -43,20 +43,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+		reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+		    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -67,7 +67,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+	        .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -91,7 +91,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+	    int seed=0)
 
         {
             var random = new Random(seed);
@@ -107,7 +107,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     // The feature values are slightly increased by adding a
 		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+		        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -138,7 +138,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From 1af15548da2c3a8c4998d01f888159969d61d5b3 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:24:53 -0700
Subject: [PATCH 29/44] Update LightGbmWithOptions.cs

---
 .../LightGbmWithOptions.cs                     | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
index 16c73ff68b..322b371ac1 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
@@ -43,7 +43,7 @@ public static void Example()
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply LightGbm multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .LightGbm(options));
+		        .LightGbm(options));
 			
 
             // Train the model.
@@ -52,20 +52,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+	        .CreateEnumerable<Prediction>(transformedTestData,
+		reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+		    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -76,7 +76,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+	        .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -100,7 +100,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+	    int seed=0)
 
         {
             var random = new Random(seed);
@@ -116,7 +116,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     // The feature values are slightly increased by adding a
 		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+		        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -147,7 +147,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From d0e1a019f4af4de36399867a0b42a8a2267f4cfb Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:28:34 -0700
Subject: [PATCH 30/44] Update MulticlassClassification.ttinclude

---
 .../MulticlassClassification.ttinclude        | 28 +++++++++----------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
index 9790c6a359..01973d367b 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
@@ -33,7 +33,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 	    // data can fit into memory, a solution is to cache the data in memory.
 	    // Caching is especially helpful when working with iterative algorithms 
 	    // which needs many data passes.
-			trainingData = mlContext.Data.Cache(trainingData);
+	    trainingData = mlContext.Data.Cache(trainingData);
 <# } #>
 
 <# if (MetaTrainer != null) { #>
@@ -44,19 +44,19 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                     // Apply <#=MetaTrainer#> multiclass meta trainer on top of
 		    // binary trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .<#=MetaTrainer#>(
-						<#=Trainer#>()));
+		        .<#=MetaTrainer#>(
+			<#=Trainer#>()));
 
 <# } else if (TrainerOptions == null) { #>
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-					    .MapValueToKey(nameof(DataPoint.Label))
+		        .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply <#=Trainer#> multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .<#=Trainer#>());
+		        .<#=Trainer#>());
 
 <# } else { #>
             // Define trainer options.
@@ -68,7 +68,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply <#=Trainer#> multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .<#=Trainer#>(options));
+		        .<#=Trainer#>(options));
 			
 <# } #>
 
@@ -78,26 +78,26 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+	        .CreateEnumerable<Prediction>(transformedTestData,
+		reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+		    $"Prediction: {p.PredictedLabel}");
 
             <#=ExpectedOutputPerInstance#>
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+	        .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -106,7 +106,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 
         <#=DataGenerationComments#>
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -122,7 +122,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                     // The feature values are slightly increased by adding a
 		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -153,7 +153,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From 092808dc5a532ed56a651a29a20866945766504a Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:30:27 -0700
Subject: [PATCH 31/44] Update NaiveBayes.cs

---
 .../MulticlassClassification/NaiveBayes.cs    | 20 +++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
index 354352b3c7..0afd943982 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
@@ -33,11 +33,11 @@ public static void Example()
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-					    .MapValueToKey(nameof(DataPoint.Label))
+		        .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply NaiveBayes multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .NaiveBayes());
+                        .NaiveBayes());
 
 
             // Train the model.
@@ -46,20 +46,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+		reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+		    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -70,7 +70,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+	        .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -96,7 +96,7 @@ public static void Example()
 	// 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
 	// zero or less are treated as false.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+	    int seed=0)
 
         {
             var random = new Random(seed);
@@ -112,7 +112,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     // The feature values are slightly increased by adding a
 		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+		        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -143,7 +143,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From e9432305836aec563082b4b0d6604e1d2b298212 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:32:44 -0700
Subject: [PATCH 32/44] Update OneVersusAll.cs

---
 .../MulticlassClassification/OneVersusAll.cs  | 20 +++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
index bab768a9d7..96ce036af9 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
@@ -30,8 +30,8 @@ public static void Example()
                     // Apply OneVersusAll multiclass meta trainer on top of
 		    // binary trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .OneVersusAll(
-						mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+	                .OneVersusAll(
+			mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
 
 
             // Train the model.
@@ -40,20 +40,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+	        .CreateEnumerable<Prediction>(transformedTestData,
+		reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+		    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -64,7 +64,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+	        .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -88,7 +88,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+	    int seed=0)
 
         {
             var random = new Random(seed);
@@ -104,7 +104,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     // The feature values are slightly increased by adding a
 		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+		        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -135,7 +135,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From 9d014e3f827d5d0ad9d6a0027040a332f623702d Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:34:22 -0700
Subject: [PATCH 33/44] Update PairwiseCoupling.cs

---
 .../PairwiseCoupling.cs                       | 22 +++++++++----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
index f933989392..6a1bae9cd3 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
@@ -28,10 +28,10 @@ public static void Example()
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply PairwiseCoupling multiclass meta trainer on top of
-					// binary trainer.
+		    // binary trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .PairwiseCoupling(
-						mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+		        .PairwiseCoupling(
+			mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
 
 
             // Train the model.
@@ -40,20 +40,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+	        .CreateEnumerable<Prediction>(transformedTestData,
+	        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+		    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -64,7 +64,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+	        .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -88,7 +88,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -104,7 +104,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     // The feature values are slightly increased by adding a
 		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+		        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -135,7 +135,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From 407c4d8df26691d715c0acc93fac0c1b22e5238f Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:37:23 -0700
Subject: [PATCH 34/44] Update SdcaMaximumEntropy.cs

---
 .../SdcaMaximumEntropy.cs                     | 22 +++++++++----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
index be4cf22296..2c94c858df 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
@@ -29,17 +29,17 @@ public static void Example()
 	    // data can fit into memory, a solution is to cache the data in memory.
 	    // Caching is especially helpful when working with iterative algorithms 
 	    // which needs many data passes.
-			trainingData = mlContext.Data.Cache(trainingData);
+	    trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-					    .MapValueToKey(nameof(DataPoint.Label))
+		    .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply SdcaMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .SdcaMaximumEntropy());
+		    .SdcaMaximumEntropy());
 
 
             // Train the model.
@@ -48,20 +48,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+	        .CreateEnumerable<Prediction>(transformedTestData,
+	        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+	            $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -72,7 +72,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+	        .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -95,7 +95,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+	    int seed=0)
 
         {
             var random = new Random(seed);
@@ -111,7 +111,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     // The feature values are slightly increased by adding a
 		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+		        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -142,7 +142,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From b4c95757cacdfea0d106c3690867bc1d39c063e3 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:39:05 -0700
Subject: [PATCH 35/44] Update SdcaMaximumEntropy.cs

---
 .../Trainers/MulticlassClassification/SdcaMaximumEntropy.cs   | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
index 2c94c858df..72981bc20a 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
@@ -35,11 +35,11 @@ public static void Example()
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-		    .MapValueToKey(nameof(DataPoint.Label))
+		        .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply SdcaMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-		    .SdcaMaximumEntropy());
+		        .SdcaMaximumEntropy());
 
 
             // Train the model.

From 053c85bbb0cb0f8b1cda57336b7d25c128366241 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:41:08 -0700
Subject: [PATCH 36/44] Update SdcaMaximumEntropyWithOptions.cs

---
 .../SdcaMaximumEntropyWithOptions.cs          | 20 +++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
index 0501abb458..891c9cfc3f 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
@@ -30,7 +30,7 @@ public static void Example()
 	    // data can fit into memory, a solution is to cache the data in memory.
 	    // Caching is especially helpful when working with iterative algorithms 
 	    // which needs many data passes.
-			trainingData = mlContext.Data.Cache(trainingData);
+	    trainingData = mlContext.Data.Cache(trainingData);
 
             // Define trainer options.
             var options = new SdcaMaximumEntropyMulticlassTrainer.Options
@@ -47,7 +47,7 @@ public static void Example()
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply SdcaMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .SdcaMaximumEntropy(options));
+	                .SdcaMaximumEntropy(options));
 			
 
             // Train the model.
@@ -56,20 +56,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+	        .CreateEnumerable<Prediction>(transformedTestData,
+		reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+		    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -80,7 +80,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+	        .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -104,7 +104,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+	    int seed=0)
 
         {
             var random = new Random(seed);
@@ -120,7 +120,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     // The feature values are slightly increased by adding a
 		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+		        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -151,7 +151,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From 0d2baa69f0689eb3dfeb59078496d2777fdec331 Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:42:49 -0700
Subject: [PATCH 37/44] Update SdcaNonCalibrated.cs

---
 .../SdcaNonCalibrated.cs                      | 20 +++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index cf33d4ae1b..e67ea997c4 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -35,11 +35,11 @@ public static void Example()
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-					    .MapValueToKey(nameof(DataPoint.Label))
+                        .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply SdcaNonCalibrated multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .SdcaNonCalibrated());
+		        .SdcaNonCalibrated());
 
 
             // Train the model.
@@ -48,20 +48,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-			    reuseRowObject: false).ToList();
+	        .CreateEnumerable<Prediction>(transformedTestData,
+		reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+		    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -72,7 +72,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+	        .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -96,7 +96,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+	    int seed=0)
 
         {
             var random = new Random(seed);
@@ -112,7 +112,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     // The feature values are slightly increased by adding a
 		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+		        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -143,7 +143,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From 6e410cc2e4ff65774e1ced31fc46dbb2311d57ff Mon Sep 17 00:00:00 2001
From: Sierra Lee <sierrajadelee@gmail.com>
Date: Mon, 1 Jul 2019 14:44:56 -0700
Subject: [PATCH 38/44] Update SdcaNonCalibratedWithOptions.cs

---
 .../SdcaNonCalibratedWithOptions.cs            | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
index 44e40ff6df..f8a945127c 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
@@ -47,7 +47,7 @@ public static void Example()
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply SdcaNonCalibrated multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-					    .SdcaNonCalibrated(options));
+		        .SdcaNonCalibrated(options));
 			
 
             // Train the model.
@@ -56,20 +56,20 @@ public static void Example()
             // Create testing data. Use different random seed to make it different
 	    // from training data.
             var testData = mlContext.Data
-			    .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-			    .CreateEnumerable<Prediction>(transformedTestData,
-				reuseRowObject: false).ToList();
+	        .CreateEnumerable<Prediction>(transformedTestData,
+		reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-				    $"Prediction: {p.PredictedLabel}");
+		    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -80,7 +80,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-			    .Evaluate(transformedTestData);
+	        .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -104,7 +104,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-		    int seed=0)
+	    int seed=0)
 
         {
             var random = new Random(seed);
@@ -120,7 +120,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     // The feature values are slightly increased by adding a
 		    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-					    .Select(x => randomFloat() + label * 0.2f).ToArray()
+		        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -151,7 +151,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-			    $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From 85ed0e6ce2ed95e61cd08455d6ec4828ba3d6e9c Mon Sep 17 00:00:00 2001
From: sierralee51 <36902786+sierralee51@users.noreply.github.com>
Date: Mon, 1 Jul 2019 23:24:52 -0700
Subject: [PATCH 39/44] fixed tabbing issue

---
 .../LbfgsMaximumEntropy.cs                    | 24 ++++-----
 .../LbfgsMaximumEntropyWithOptions.cs         | 22 ++++----
 .../MulticlassClassification/LightGbm.cs      | 32 ++++++------
 .../LightGbmWithOptions.cs                    | 34 ++++++------
 .../MulticlassClassification.ttinclude        | 52 +++++++++----------
 .../MulticlassClassification/NaiveBayes.cs    | 47 +++++++++--------
 .../MulticlassClassification/NaiveBayes.tt    | 10 ++--
 .../MulticlassClassification/OneVersusAll.cs  | 37 ++++++-------
 .../PairwiseCoupling.cs                       | 32 ++++++------
 .../SdcaMaximumEntropy.cs                     | 46 ++++++++--------
 .../SdcaMaximumEntropyWithOptions.cs          | 48 ++++++++---------
 .../SdcaNonCalibrated.cs                      | 42 +++++++--------
 .../SdcaNonCalibratedWithOptions.cs           | 49 ++++++++---------
 13 files changed, 239 insertions(+), 236 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index d24869e6a4..a5b27c2693 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -11,34 +11,34 @@ public static class LbfgsMaximumEntropy
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-		        .MapValueToKey(nameof(DataPoint.Label))
+                .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply LbfgsMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-	                .LbfgsMaximumEntropy());
+                    .LbfgsMaximumEntropy());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
                 .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -48,12 +48,12 @@ public static void Example()
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
                 .CreateEnumerable<Prediction>(transformedTestData,
-	        reuseRowObject: false).ToList();
+            reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-	            $"Prediction: {p.PredictedLabel}");
+                $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -88,7 +88,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, 
-	    int seed=0)
+        int seed=0)
 
         {
             var random = new Random(seed);
@@ -102,9 +102,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-	                .Select(x => randomFloat() + label * 0.2f).ToArray()
+                    .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
index 89728c59f9..08bfd399b7 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
@@ -12,16 +12,16 @@ public static class LbfgsMaximumEntropyWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define trainer options.
@@ -34,18 +34,18 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-	            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply LbfgsMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-	                .LbfgsMaximumEntropy(options));
-			
+                    .LbfgsMaximumEntropy(options));
+            
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
                 .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -55,7 +55,7 @@ public static void Example()
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
                 .CreateEnumerable<Prediction>(transformedTestData,
-	        reuseRowObject: false).ToList();
+            reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
@@ -109,9 +109,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-		        .Select(x => randomFloat() + label * 0.2f).ToArray()
+                .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
index 0d898b6790..91033d8969 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
@@ -14,36 +14,36 @@ public static class LightGbm
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-		        .MapValueToKey(nameof(DataPoint.Label))
+                .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply LightGbm multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-			.LightGbm());
+            .LightGbm());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
-	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
@@ -51,12 +51,12 @@ public static void Example()
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
                 .CreateEnumerable<Prediction>(transformedTestData,
-		reuseRowObject: false).ToList();
+        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-		    $"Prediction: {p.PredictedLabel}");
+            $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -67,7 +67,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-	        .Evaluate(transformedTestData);
+            .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -91,7 +91,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-	    int seed=0)
+        int seed=0)
 
         {
             var random = new Random(seed);
@@ -105,16 +105,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-		        .Select(x => randomFloat() + label * 0.2f).ToArray()
+                .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-	// such examples.
+    // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -138,7 +138,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
index 322b371ac1..e68fe66b6e 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
@@ -15,16 +15,16 @@ public static class LightGbmWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define trainer options.
@@ -39,33 +39,33 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-	            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply LightGbm multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-		        .LightGbm(options));
-			
+                .LightGbm(options));
+            
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
-	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-	        .CreateEnumerable<Prediction>(transformedTestData,
-		reuseRowObject: false).ToList();
+            .CreateEnumerable<Prediction>(transformedTestData,
+        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-		    $"Prediction: {p.PredictedLabel}");
+            $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -76,7 +76,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-	        .Evaluate(transformedTestData);
+            .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -100,7 +100,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-	    int seed=0)
+        int seed=0)
 
         {
             var random = new Random(seed);
@@ -114,9 +114,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-		        .Select(x => randomFloat() + label * 0.2f).ToArray()
+                .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -147,7 +147,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
index 01973d367b..21006d12f2 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
@@ -14,26 +14,26 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 <# if (CacheData) { #>
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-	    // data set from a file and accesses it many times, it can be slow due
-	    // to expensive featurization and disk operations. When the considered
-	    // data can fit into memory, a solution is to cache the data in memory.
-	    // Caching is especially helpful when working with iterative algorithms 
-	    // which needs many data passes.
-	    trainingData = mlContext.Data.Cache(trainingData);
+        // data set from a file and accesses it many times, it can be slow due
+        // to expensive featurization and disk operations. When the considered
+        // data can fit into memory, a solution is to cache the data in memory.
+        // Caching is especially helpful when working with iterative algorithms 
+        // which needs many data passes.
+        trainingData = mlContext.Data.Cache(trainingData);
 <# } #>
 
 <# if (MetaTrainer != null) { #>
@@ -42,21 +42,21 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply <#=MetaTrainer#> multiclass meta trainer on top of
-		    // binary trainer.
+            // binary trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-		        .<#=MetaTrainer#>(
-			<#=Trainer#>()));
+                .<#=MetaTrainer#>(
+            <#=Trainer#>()));
 
 <# } else if (TrainerOptions == null) { #>
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-		        .MapValueToKey(nameof(DataPoint.Label))
+                .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply <#=Trainer#> multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-		        .<#=Trainer#>());
+                .<#=Trainer#>());
 
 <# } else { #>
             // Define trainer options.
@@ -68,36 +68,36 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply <#=Trainer#> multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-		        .<#=Trainer#>(options));
-			
+                .<#=Trainer#>(options));
+            
 <# } #>
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
-	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-	        .CreateEnumerable<Prediction>(transformedTestData,
-		reuseRowObject: false).ToList();
+            .CreateEnumerable<Prediction>(transformedTestData,
+        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-		    $"Prediction: {p.PredictedLabel}");
+            $"Prediction: {p.PredictedLabel}");
 
             <#=ExpectedOutputPerInstance#>
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-	        .Evaluate(transformedTestData);
+            .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -120,7 +120,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
                         .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -129,7 +129,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-	// such examples.
+    // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -153,7 +153,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
index 0afd943982..ef62fac759 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
@@ -10,56 +10,56 @@ public static class NaiveBayes
     {
         // Naive Bayes classifier is based on Bayes' theorem. 
         // It assumes independence among the presence of features in a class even
-	// though they may be dependent on each other. It is  a multi-class trainer
-	// that accepts binary feature values of type float, i.e., feature values
-	// are either true or false. Specifically a feature value greater than zero
-	// is treated as true, zero or less is treated as false.
+        // though they may be dependent on each other. It is  a multi-class trainer
+    // that accepts binary feature values of type float, i.e., feature values
+    // are either true or false. Specifically a feature value greater than zero
+    // is treated as true, zero or less is treated as false.
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-		        .MapValueToKey(nameof(DataPoint.Label))
+                .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply NaiveBayes multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-                        .NaiveBayes());
+                .NaiveBayes());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
-                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-                .CreateEnumerable<Prediction>(transformedTestData,
-		reuseRowObject: false).ToList();
+            .CreateEnumerable<Prediction>(transformedTestData,
+        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-		    $"Prediction: {p.PredictedLabel}");
+            $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -70,7 +70,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-	        .Evaluate(transformedTestData);
+            .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -93,10 +93,10 @@ public static void Example()
 
         
         // Generates random uniform doubles in [-0.5, 0.5) range with labels
-	// 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
-	// zero or less are treated as false.
+    // 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
+    // zero or less are treated as false.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-	    int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -110,16 +110,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-		        .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-	// such examples.
+    // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -143,9 +143,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
 }
+
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
index ca83a2b378..2e15f6af2e 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
@@ -11,14 +11,14 @@ string Comments= @"
         // Naive Bayes classifier is based on Bayes' theorem. 
         // It assumes independence among the presence of features in a class even
         // though they may be dependent on each other. It is  a multi-class trainer
-	// that accepts binary feature values of type float, i.e., feature values
-	// are either true or false. Specifically a feature value greater than zero
-	// is treated as true, zero or less is treated as false.";
+    // that accepts binary feature values of type float, i.e., feature values
+    // are either true or false. Specifically a feature value greater than zero
+    // is treated as true, zero or less is treated as false.";
 
 string DataGenerationComments= @"
         // Generates random uniform doubles in [-0.5, 0.5) range with labels
-	// 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
-	// zero or less are treated as false.";
+    // 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
+    // zero or less are treated as false.";
 
 string ExpectedOutputPerInstance= @"// Expected output:
             //   Label: 1, Prediction: 1
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
index 96ce036af9..357cb28a98 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
@@ -11,16 +11,16 @@ public static class OneVersusAll
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
@@ -28,32 +28,32 @@ public static void Example()
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply OneVersusAll multiclass meta trainer on top of
-		    // binary trainer.
+            // binary trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-	                .OneVersusAll(
-			mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+                .OneVersusAll(
+            mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
-	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-	        .CreateEnumerable<Prediction>(transformedTestData,
-		reuseRowObject: false).ToList();
+            .CreateEnumerable<Prediction>(transformedTestData,
+        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-		    $"Prediction: {p.PredictedLabel}");
+            $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -64,7 +64,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-	        .Evaluate(transformedTestData);
+            .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -88,7 +88,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-	    int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -102,16 +102,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-		        .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-        // such examples.
+    // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -135,9 +135,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
 }
+
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
index 6a1bae9cd3..1eee975e42 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
@@ -11,16 +11,16 @@ public static class PairwiseCoupling
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
@@ -28,32 +28,32 @@ public static void Example()
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply PairwiseCoupling multiclass meta trainer on top of
-		    // binary trainer.
+            // binary trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-		        .PairwiseCoupling(
-			mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+                .PairwiseCoupling(
+            mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
-	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-	        .CreateEnumerable<Prediction>(transformedTestData,
-	        reuseRowObject: false).ToList();
+            .CreateEnumerable<Prediction>(transformedTestData,
+            reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-		    $"Prediction: {p.PredictedLabel}");
+            $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -64,7 +64,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-	        .Evaluate(transformedTestData);
+            .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -102,9 +102,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-		        .Select(x => randomFloat() + label * 0.2f).ToArray()
+                .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -135,7 +135,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
index 72981bc20a..b2175bb7bc 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
@@ -11,57 +11,57 @@ public static class SdcaMaximumEntropy
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-	    // data set from a file and accesses it many times, it can be slow due
-	    // to expensive featurization and disk operations. When the considered
-	    // data can fit into memory, a solution is to cache the data in memory.
-	    // Caching is especially helpful when working with iterative algorithms 
-	    // which needs many data passes.
-	    trainingData = mlContext.Data.Cache(trainingData);
+        // data set from a file and accesses it many times, it can be slow due
+        // to expensive featurization and disk operations. When the considered
+        // data can fit into memory, a solution is to cache the data in memory.
+        // Caching is especially helpful when working with iterative algorithms 
+        // which needs many data passes.
+        trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
                     // Convert the string labels into key types.
                     mlContext.Transforms.Conversion
-		        .MapValueToKey(nameof(DataPoint.Label))
+                .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply SdcaMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-		        .SdcaMaximumEntropy());
+                .SdcaMaximumEntropy());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
-	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-	        .CreateEnumerable<Prediction>(transformedTestData,
-	        reuseRowObject: false).ToList();
+            .CreateEnumerable<Prediction>(transformedTestData,
+            reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-	            $"Prediction: {p.PredictedLabel}");
+                $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -72,7 +72,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-	        .Evaluate(transformedTestData);
+            .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -95,7 +95,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-	    int seed=0)
+        int seed=0)
 
         {
             var random = new Random(seed);
@@ -109,16 +109,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-		        .Select(x => randomFloat() + label * 0.2f).ToArray()
+                .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-	// such examples.
+    // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -142,7 +142,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
index 891c9cfc3f..f7fd099a98 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
@@ -12,25 +12,25 @@ public static class SdcaMaximumEntropyWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-	    // data set from a file and accesses it many times, it can be slow due
-	    // to expensive featurization and disk operations. When the considered
-	    // data can fit into memory, a solution is to cache the data in memory.
-	    // Caching is especially helpful when working with iterative algorithms 
-	    // which needs many data passes.
-	    trainingData = mlContext.Data.Cache(trainingData);
+        // data set from a file and accesses it many times, it can be slow due
+        // to expensive featurization and disk operations. When the considered
+        // data can fit into memory, a solution is to cache the data in memory.
+        // Caching is especially helpful when working with iterative algorithms 
+        // which needs many data passes.
+        trainingData = mlContext.Data.Cache(trainingData);
 
             // Define trainer options.
             var options = new SdcaMaximumEntropyMulticlassTrainer.Options
@@ -43,33 +43,33 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-	            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply SdcaMaximumEntropy multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-	                .SdcaMaximumEntropy(options));
-			
+                    .SdcaMaximumEntropy(options));
+            
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
-	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-	        .CreateEnumerable<Prediction>(transformedTestData,
-		reuseRowObject: false).ToList();
+            .CreateEnumerable<Prediction>(transformedTestData,
+        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-		    $"Prediction: {p.PredictedLabel}");
+            $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -80,7 +80,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-	        .Evaluate(transformedTestData);
+            .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -104,7 +104,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-	    int seed=0)
+        int seed=0)
 
         {
             var random = new Random(seed);
@@ -118,16 +118,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-		        .Select(x => randomFloat() + label * 0.2f).ToArray()
+                .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-	// such examples.
+    // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -151,7 +151,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index e67ea997c4..83415a1e6c 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -11,25 +11,25 @@ public static class SdcaNonCalibrated
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-	    // data set from a file and accesses it many times, it can be slow due
-	    // to expensive featurization and disk operations. When the considered
-	    // data can fit into memory, a solution is to cache the data in memory.
-	    // Caching is especially helpful when working with iterative algorithms 
-	    // which needs many data passes.
-	    trainingData = mlContext.Data.Cache(trainingData);
+        // data set from a file and accesses it many times, it can be slow due
+        // to expensive featurization and disk operations. When the considered
+        // data can fit into memory, a solution is to cache the data in memory.
+        // Caching is especially helpful when working with iterative algorithms 
+        // which needs many data passes.
+        trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
@@ -39,29 +39,29 @@ public static void Example()
 
                     // Apply SdcaNonCalibrated multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-		        .SdcaNonCalibrated());
+                .SdcaNonCalibrated());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
-	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-	        .CreateEnumerable<Prediction>(transformedTestData,
-		reuseRowObject: false).ToList();
+            .CreateEnumerable<Prediction>(transformedTestData,
+        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-		    $"Prediction: {p.PredictedLabel}");
+            $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -72,7 +72,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-	        .Evaluate(transformedTestData);
+            .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -96,7 +96,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-	    int seed=0)
+        int seed=0)
 
         {
             var random = new Random(seed);
@@ -110,9 +110,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-		        .Select(x => randomFloat() + label * 0.2f).ToArray()
+                .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -143,7 +143,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
index f8a945127c..7d842848b2 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
@@ -12,25 +12,25 @@ public static class SdcaNonCalibratedWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-	    // exception tracking and logging, as a catalog of available operations
-	    // and as the source of randomness. Setting the seed to a fixed number
-	    // in this example to make outputs deterministic.
+        // exception tracking and logging, as a catalog of available operations
+        // and as the source of randomness. Setting the seed to a fixed number
+        // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-	    // consumable by ML.NET API.
+        // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-	    // data set from a file and accesses it many times, it can be slow due
-	    // to expensive featurization and disk operations. When the considered
-	    // data can fit into memory, a solution is to cache the data in memory.
-	    // Caching is especially helpful when working with iterative algorithms 
-	    // which needs many data passes.
-	    trainingData = mlContext.Data.Cache(trainingData);
+        // data set from a file and accesses it many times, it can be slow due
+        // to expensive featurization and disk operations. When the considered
+        // data can fit into memory, a solution is to cache the data in memory.
+        // Caching is especially helpful when working with iterative algorithms 
+        // which needs many data passes.
+        trainingData = mlContext.Data.Cache(trainingData);
 
             // Define trainer options.
             var options = new SdcaNonCalibratedMulticlassTrainer.Options
@@ -43,33 +43,33 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-	            // Convert the string labels into key types.
+                    // Convert the string labels into key types.
                     mlContext.Transforms.Conversion.MapValueToKey("Label")
                     // Apply SdcaNonCalibrated multiclass trainer.
                     .Append(mlContext.MulticlassClassification.Trainers
-		        .SdcaNonCalibrated(options));
-			
+                .SdcaNonCalibrated(options));
+            
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-	    // from training data.
+        // from training data.
             var testData = mlContext.Data
-	        .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-	        .CreateEnumerable<Prediction>(transformedTestData,
-		reuseRowObject: false).ToList();
+            .CreateEnumerable<Prediction>(transformedTestData,
+        reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-		    $"Prediction: {p.PredictedLabel}");
+            $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -80,7 +80,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-	        .Evaluate(transformedTestData);
+            .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -104,7 +104,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-	    int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -118,16 +118,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-		    // constant multiple of label.
+            // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-		        .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-	// such examples.
+    // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -151,9 +151,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-	        $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
 }
+

From f51c1478b8e84ead4ea6683a9c7fca2bf8ccbffd Mon Sep 17 00:00:00 2001
From: sierralee51 <36902786+sierralee51@users.noreply.github.com>
Date: Mon, 1 Jul 2019 23:40:27 -0700
Subject: [PATCH 40/44] fixed indentations

---
 .../LbfgsMaximumEntropy.cs                    | 34 +++++-----
 .../LbfgsMaximumEntropyWithOptions.cs         | 27 ++++----
 .../MulticlassClassification/LightGbm.cs      | 40 ++++++------
 .../LightGbmWithOptions.cs                    | 37 ++++++-----
 .../MulticlassClassification.ttinclude        | 65 +++++++++----------
 5 files changed, 102 insertions(+), 101 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index a5b27c2693..3b30065324 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -11,34 +11,33 @@ public static class LbfgsMaximumEntropy
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-
-                    // Apply LbfgsMaximumEntropy multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
-                    .LbfgsMaximumEntropy());
+            // Apply LbfgsMaximumEntropy multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
+                .LbfgsMaximumEntropy());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
                 .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -48,12 +47,12 @@ public static void Example()
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
                 .CreateEnumerable<Prediction>(transformedTestData,
-            reuseRowObject: false).ToList();
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-                $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -87,8 +86,8 @@ public static void Example()
 
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
-        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, 
-        int seed=0)
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -102,9 +101,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                    .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -141,3 +140,4 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
         }
     }
 }
+
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
index 08bfd399b7..af7922618d 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
@@ -12,16 +12,16 @@ public static class LbfgsMaximumEntropyWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define trainer options.
@@ -34,18 +34,18 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-                // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply LbfgsMaximumEntropy multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
-                    .LbfgsMaximumEntropy(options));
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion.MapValueToKey("Label")
+            // Apply LbfgsMaximumEntropy multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
+                .LbfgsMaximumEntropy(options));
             
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
                 .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
@@ -55,7 +55,7 @@ public static void Example()
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
                 .CreateEnumerable<Prediction>(transformedTestData,
-            reuseRowObject: false).ToList();
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
@@ -109,9 +109,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -148,3 +148,4 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
         }
     }
 }
+
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
index 91033d8969..7992abe2bf 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
@@ -14,36 +14,35 @@ public static class LightGbm
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-
-                    // Apply LightGbm multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
-            .LightGbm());
+            // Apply LightGbm multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
+                .LightGbm());
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
-            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
@@ -51,12 +50,12 @@ public static void Example()
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
                 .CreateEnumerable<Prediction>(transformedTestData,
-        reuseRowObject: false).ToList();
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-            $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -67,7 +66,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-            .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -91,7 +90,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-        int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -105,16 +104,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-    // such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -138,9 +137,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
 }
+
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
index e68fe66b6e..e814030061 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
@@ -15,16 +15,16 @@ public static class LightGbmWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define trainer options.
@@ -39,10 +39,10 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-                // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply LightGbm multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion.MapValueToKey("Label")
+            // Apply LightGbm multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
                 .LightGbm(options));
             
 
@@ -50,22 +50,22 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
-            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-            .CreateEnumerable<Prediction>(transformedTestData,
-        reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-            $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -76,7 +76,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-            .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -100,7 +100,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-        int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -114,9 +114,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -147,9 +147,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
 }
+
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
index 21006d12f2..7ccc793feb 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
@@ -14,48 +14,47 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 <# if (CacheData) { #>
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-        // data set from a file and accesses it many times, it can be slow due
-        // to expensive featurization and disk operations. When the considered
-        // data can fit into memory, a solution is to cache the data in memory.
-        // Caching is especially helpful when working with iterative algorithms 
-        // which needs many data passes.
-        trainingData = mlContext.Data.Cache(trainingData);
+            // data set from a file and accesses it many times, it can be slow due
+            // to expensive featurization and disk operations. When the considered
+            // data can fit into memory, a solution is to cache the data in memory.
+            // Caching is especially helpful when working with iterative algorithms 
+            // which needs many data passes.
+            trainingData = mlContext.Data.Cache(trainingData);
 <# } #>
 
 <# if (MetaTrainer != null) { #>
             // Define the trainer.
             var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply <#=MetaTrainer#> multiclass meta trainer on top of
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion.MapValueToKey("Label")
+            // Apply <#=MetaTrainer#> multiclass meta trainer on top of
             // binary trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
+                .Append(mlContext.MulticlassClassification.Trainers
                 .<#=MetaTrainer#>(
-            <#=Trainer#>()));
+                <#=Trainer#>()));
 
 <# } else if (TrainerOptions == null) { #>
             // Define the trainer.
             var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-
-                    // Apply <#=Trainer#> multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
+            // Apply <#=Trainer#> multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
                 .<#=Trainer#>());
 
 <# } else { #>
@@ -64,10 +63,10 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 
             // Define the trainer.
             var pipeline = 
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply <#=Trainer#> multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion.MapValueToKey("Label")
+            // Apply <#=Trainer#> multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
                 .<#=Trainer#>(options));
             
 <# } #>
@@ -76,28 +75,28 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
-            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-            .CreateEnumerable<Prediction>(transformedTestData,
-        reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-            $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             <#=ExpectedOutputPerInstance#>
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-            .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -120,7 +119,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
                         .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -129,7 +128,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-    // such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -153,7 +152,7 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From 1e9a3d881bd1a7ee7c303cb56aa77c1cd6399b0e Mon Sep 17 00:00:00 2001
From: sierralee51 <36902786+sierralee51@users.noreply.github.com>
Date: Mon, 1 Jul 2019 23:49:22 -0700
Subject: [PATCH 41/44] aligned comments

---
 .../LogLossPerClass.cs                        |  3 +-
 .../MulticlassClassification/NaiveBayes.cs    | 45 ++++++++--------
 .../MulticlassClassification/NaiveBayes.tt    | 10 ++--
 .../MulticlassClassification/OneVersusAll.cs  | 36 ++++++-------
 .../PairwiseCoupling.cs                       | 37 ++++++-------
 .../SdcaMaximumEntropy.cs                     | 52 +++++++++---------
 .../SdcaMaximumEntropyWithOptions.cs          | 53 ++++++++++---------
 .../SdcaNonCalibrated.cs                      | 52 +++++++++---------
 .../SdcaNonCalibratedWithOptions.cs           | 46 ++++++++--------
 9 files changed, 168 insertions(+), 166 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs
index c875616ada..02550a5bb1 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs
@@ -30,7 +30,8 @@ public static void Example()
                         .MapValueToKey(nameof(DataPoint.Label))
 
                     // Apply a multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers.LightGbm());
+                        .Append(mlContext.MulticlassClassification.Trainers
+                        .LightGbm());
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
index ef62fac759..33272b2fb0 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
@@ -11,32 +11,31 @@ public static class NaiveBayes
         // Naive Bayes classifier is based on Bayes' theorem. 
         // It assumes independence among the presence of features in a class even
         // though they may be dependent on each other. It is  a multi-class trainer
-    // that accepts binary feature values of type float, i.e., feature values
-    // are either true or false. Specifically a feature value greater than zero
-    // is treated as true, zero or less is treated as false.
+        // that accepts binary feature values of type float, i.e., feature values
+        // are either true or false. Specifically a feature value greater than zero
+        // is treated as true, zero or less is treated as false.
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-
-                    // Apply NaiveBayes multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
+            // Apply NaiveBayes multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
                 .NaiveBayes());
 
 
@@ -44,22 +43,22 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
-            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-            .CreateEnumerable<Prediction>(transformedTestData,
-        reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-            $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -70,7 +69,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-            .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -93,8 +92,8 @@ public static void Example()
 
         
         // Generates random uniform doubles in [-0.5, 0.5) range with labels
-    // 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
-    // zero or less are treated as false.
+        // 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
+        // zero or less are treated as false.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
             int seed=0)
 
@@ -110,7 +109,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
                         .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -119,7 +118,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-    // such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -143,7 +142,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
index 2e15f6af2e..26e0162b12 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.tt
@@ -11,14 +11,14 @@ string Comments= @"
         // Naive Bayes classifier is based on Bayes' theorem. 
         // It assumes independence among the presence of features in a class even
         // though they may be dependent on each other. It is  a multi-class trainer
-    // that accepts binary feature values of type float, i.e., feature values
-    // are either true or false. Specifically a feature value greater than zero
-    // is treated as true, zero or less is treated as false.";
+        // that accepts binary feature values of type float, i.e., feature values
+        // are either true or false. Specifically a feature value greater than zero
+        // is treated as true, zero or less is treated as false.";
 
 string DataGenerationComments= @"
         // Generates random uniform doubles in [-0.5, 0.5) range with labels
-    // 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
-    // zero or less are treated as false.";
+        // 1, 2 or 3. For NaiveBayes values greater than zero are treated as true,
+        // zero or less are treated as false.";
 
 string ExpectedOutputPerInstance= @"// Expected output:
             //   Label: 1, Prediction: 1
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
index 357cb28a98..044cfcd3fb 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
@@ -11,49 +11,49 @@ public static class OneVersusAll
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply OneVersusAll multiclass meta trainer on top of
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion.MapValueToKey("Label")
+            // Apply OneVersusAll multiclass meta trainer on top of
             // binary trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
+                .Append(mlContext.MulticlassClassification.Trainers
                 .OneVersusAll(
-            mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+                mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
-            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-            .CreateEnumerable<Prediction>(transformedTestData,
-        reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-            $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -64,7 +64,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-            .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -102,7 +102,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
                         .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -111,7 +111,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-    // such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -135,7 +135,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
index 1eee975e42..a2b518bc1a 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
@@ -11,49 +11,49 @@ public static class PairwiseCoupling
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // Define the trainer.
             var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply PairwiseCoupling multiclass meta trainer on top of
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion.MapValueToKey("Label")
+            // Apply PairwiseCoupling multiclass meta trainer on top of
             // binary trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
+                .Append(mlContext.MulticlassClassification.Trainers
                 .PairwiseCoupling(
-            mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
+                mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
 
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
-            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-            .CreateEnumerable<Prediction>(transformedTestData,
-            reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-            $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -64,7 +64,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-            .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -102,9 +102,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -135,9 +135,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
 }
+
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
index b2175bb7bc..6521c0e5e1 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
@@ -11,34 +11,33 @@ public static class SdcaMaximumEntropy
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-        // data set from a file and accesses it many times, it can be slow due
-        // to expensive featurization and disk operations. When the considered
-        // data can fit into memory, a solution is to cache the data in memory.
-        // Caching is especially helpful when working with iterative algorithms 
-        // which needs many data passes.
-        trainingData = mlContext.Data.Cache(trainingData);
+            // data set from a file and accesses it many times, it can be slow due
+            // to expensive featurization and disk operations. When the considered
+            // data can fit into memory, a solution is to cache the data in memory.
+            // Caching is especially helpful when working with iterative algorithms 
+            // which needs many data passes.
+            trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-
-                    // Apply SdcaMaximumEntropy multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
+            // Apply SdcaMaximumEntropy multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaMaximumEntropy());
 
 
@@ -46,22 +45,22 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
-            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-            .CreateEnumerable<Prediction>(transformedTestData,
-            reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-                $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -72,7 +71,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-            .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -95,7 +94,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-        int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -109,16 +108,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-    // such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -142,9 +141,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
 }
+
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
index f7fd099a98..b0a61c0061 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
@@ -12,25 +12,25 @@ public static class SdcaMaximumEntropyWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-        // data set from a file and accesses it many times, it can be slow due
-        // to expensive featurization and disk operations. When the considered
-        // data can fit into memory, a solution is to cache the data in memory.
-        // Caching is especially helpful when working with iterative algorithms 
-        // which needs many data passes.
-        trainingData = mlContext.Data.Cache(trainingData);
+            // data set from a file and accesses it many times, it can be slow due
+            // to expensive featurization and disk operations. When the considered
+            // data can fit into memory, a solution is to cache the data in memory.
+            // Caching is especially helpful when working with iterative algorithms 
+            // which needs many data passes.
+            trainingData = mlContext.Data.Cache(trainingData);
 
             // Define trainer options.
             var options = new SdcaMaximumEntropyMulticlassTrainer.Options
@@ -43,33 +43,33 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-                // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply SdcaMaximumEntropy multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
-                    .SdcaMaximumEntropy(options));
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion.MapValueToKey("Label")
+            // Apply SdcaMaximumEntropy multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
+                .SdcaMaximumEntropy(options));
             
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
-            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-            .CreateEnumerable<Prediction>(transformedTestData,
-        reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-            $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -80,7 +80,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-            .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -104,7 +104,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-        int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -118,16 +118,16 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-    // such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -151,9 +151,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
 }
+
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index 83415a1e6c..4e4a612428 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -11,34 +11,33 @@ public static class SdcaNonCalibrated
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-        // data set from a file and accesses it many times, it can be slow due
-        // to expensive featurization and disk operations. When the considered
-        // data can fit into memory, a solution is to cache the data in memory.
-        // Caching is especially helpful when working with iterative algorithms 
-        // which needs many data passes.
-        trainingData = mlContext.Data.Cache(trainingData);
+            // data set from a file and accesses it many times, it can be slow due
+            // to expensive featurization and disk operations. When the considered
+            // data can fit into memory, a solution is to cache the data in memory.
+            // Caching is especially helpful when working with iterative algorithms 
+            // which needs many data passes.
+            trainingData = mlContext.Data.Cache(trainingData);
 
             // Define the trainer.
             var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion
-                        .MapValueToKey(nameof(DataPoint.Label))
-
-                    // Apply SdcaNonCalibrated multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion
+                .MapValueToKey(nameof(DataPoint.Label))
+            // Apply SdcaNonCalibrated multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaNonCalibrated());
 
 
@@ -46,22 +45,22 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
-            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-            .CreateEnumerable<Prediction>(transformedTestData,
-        reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-            $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -72,7 +71,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-            .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -96,7 +95,7 @@ public static void Example()
         // Generates random uniform doubles in [-0.5, 0.5)
         // range with labels 1, 2 or 3.
         private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
-        int seed=0)
+            int seed=0)
 
         {
             var random = new Random(seed);
@@ -110,9 +109,9 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
-                .Select(x => randomFloat() + label * 0.2f).ToArray()
+                        .Select(x => randomFloat() + label * 0.2f).ToArray()
 
                 };
             }
@@ -143,9 +142,10 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }
     }
 }
+
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
index 7d842848b2..d86bd7d1c5 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
@@ -12,25 +12,25 @@ public static class SdcaNonCalibratedWithOptions
         public static void Example()
         {
             // Create a new context for ML.NET operations. It can be used for
-        // exception tracking and logging, as a catalog of available operations
-        // and as the source of randomness. Setting the seed to a fixed number
-        // in this example to make outputs deterministic.
+            // exception tracking and logging, as a catalog of available operations
+            // and as the source of randomness. Setting the seed to a fixed number
+            // in this example to make outputs deterministic.
             var mlContext = new MLContext(seed: 0);
 
             // Create a list of training data points.
             var dataPoints = GenerateRandomDataPoints(1000);
 
             // Convert the list of data points to an IDataView object, which is
-        // consumable by ML.NET API.
+            // consumable by ML.NET API.
             var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
 
             // ML.NET doesn't cache data set by default. Therefore, if one reads a
-        // data set from a file and accesses it many times, it can be slow due
-        // to expensive featurization and disk operations. When the considered
-        // data can fit into memory, a solution is to cache the data in memory.
-        // Caching is especially helpful when working with iterative algorithms 
-        // which needs many data passes.
-        trainingData = mlContext.Data.Cache(trainingData);
+            // data set from a file and accesses it many times, it can be slow due
+            // to expensive featurization and disk operations. When the considered
+            // data can fit into memory, a solution is to cache the data in memory.
+            // Caching is especially helpful when working with iterative algorithms 
+            // which needs many data passes.
+            trainingData = mlContext.Data.Cache(trainingData);
 
             // Define trainer options.
             var options = new SdcaNonCalibratedMulticlassTrainer.Options
@@ -43,10 +43,10 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion.MapValueToKey("Label")
-                    // Apply SdcaNonCalibrated multiclass trainer.
-                    .Append(mlContext.MulticlassClassification.Trainers
+            // Convert the string labels into key types.
+                mlContext.Transforms.Conversion.MapValueToKey("Label")
+            // Apply SdcaNonCalibrated multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaNonCalibrated(options));
             
 
@@ -54,22 +54,22 @@ public static void Example()
             var model = pipeline.Fit(trainingData);
 
             // Create testing data. Use different random seed to make it different
-        // from training data.
+            // from training data.
             var testData = mlContext.Data
-            .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
+                .LoadFromEnumerable(GenerateRandomDataPoints(500, seed: 123));
 
             // Run the model on test data set.
             var transformedTestData = model.Transform(testData);
 
             // Convert IDataView object to a list.
             var predictions = mlContext.Data
-            .CreateEnumerable<Prediction>(transformedTestData,
-        reuseRowObject: false).ToList();
+                .CreateEnumerable<Prediction>(transformedTestData,
+                reuseRowObject: false).ToList();
 
             // Look at 5 predictions
             foreach (var p in predictions.Take(5))
                 Console.WriteLine($"Label: {p.Label}, " + 
-            $"Prediction: {p.PredictedLabel}");
+                    $"Prediction: {p.PredictedLabel}");
 
             // Expected output:
             //   Label: 1, Prediction: 1
@@ -80,7 +80,7 @@ public static void Example()
 
             // Evaluate the overall metrics
             var metrics = mlContext.MulticlassClassification
-            .Evaluate(transformedTestData);
+                .Evaluate(transformedTestData);
 
             PrintMetrics(metrics);
             
@@ -118,7 +118,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
                     Label = (uint)label,
                     // Create random features that are correlated with the label.
                     // The feature values are slightly increased by adding a
-            // constant multiple of label.
+                    // constant multiple of label.
                     Features = Enumerable.Repeat(label, 20)
                         .Select(x => randomFloat() + label * 0.2f).ToArray()
 
@@ -127,7 +127,7 @@ private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count,
         }
 
         // Example with label and 20 feature values. A data set is a collection of
-    // such examples.
+        // such examples.
         private class DataPoint
         {
             public uint Label { get; set; }
@@ -151,7 +151,7 @@ public static void PrintMetrics(MulticlassClassificationMetrics metrics)
             Console.WriteLine($"Macro Accuracy: {metrics.MacroAccuracy:F2}");
             Console.WriteLine($"Log Loss: {metrics.LogLoss:F2}");
             Console.WriteLine(
-            $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
+                $"Log Loss Reduction: {metrics.LogLossReduction:F2}\n");
 
             Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
         }

From c3d3499e219a704c09ed0ce50404b806d96bd43f Mon Sep 17 00:00:00 2001
From: sierralee51 <36902786+sierralee51@users.noreply.github.com>
Date: Tue, 2 Jul 2019 13:19:05 -0700
Subject: [PATCH 42/44] fixed some indentation and spacing issues

---
 .../LbfgsMaximumEntropy.cs                         |  4 ++--
 .../LbfgsMaximumEntropyWithOptions.cs              |  4 ++--
 .../Trainers/MulticlassClassification/LightGbm.cs  |  4 ++--
 .../LightGbmWithOptions.cs                         |  4 ++--
 .../MulticlassClassification/LogLossPerClass.cs    | 13 ++++++-------
 .../MulticlassClassification.ttinclude             | 14 +++++++-------
 .../MulticlassClassification/NaiveBayes.cs         |  4 ++--
 .../MulticlassClassification/OneVersusAll.cs       |  6 +++---
 .../MulticlassClassification/PairwiseCoupling.cs   |  6 +++---
 .../PermutationFeatureImportance.cs                | 10 +++++-----
 .../MulticlassClassification/SdcaMaximumEntropy.cs |  4 ++--
 .../SdcaMaximumEntropyWithOptions.cs               |  4 ++--
 .../MulticlassClassification/SdcaNonCalibrated.cs  |  4 ++--
 .../SdcaNonCalibratedWithOptions.cs                |  4 ++--
 14 files changed, 42 insertions(+), 43 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index 3b30065324..7670cf5f3d 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -25,10 +25,10 @@ public static void Example()
 
             // Define the trainer.
             var pipeline =
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-            // Apply LbfgsMaximumEntropy multiclass trainer.
+                // Apply LbfgsMaximumEntropy multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .LbfgsMaximumEntropy());
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
index af7922618d..d99c7e0d6d 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
@@ -34,9 +34,9 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion.MapValueToKey("Label")
-            // Apply LbfgsMaximumEntropy multiclass trainer.
+                // Apply LbfgsMaximumEntropy multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .LbfgsMaximumEntropy(options));
             
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
index 7992abe2bf..2cea68b970 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
@@ -28,10 +28,10 @@ public static void Example()
 
             // Define the trainer.
             var pipeline =
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-            // Apply LightGbm multiclass trainer.
+                // Apply LightGbm multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .LightGbm());
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
index e814030061..4664fad79a 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
@@ -39,9 +39,9 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion.MapValueToKey("Label")
-            // Apply LightGbm multiclass trainer.
+                // Apply LightGbm multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .LightGbm(options));
             
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs
index 02550a5bb1..8b1076f830 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LogLossPerClass.cs
@@ -25,13 +25,12 @@ public static void Example()
 
             // Define the trainer.
             var pipeline =
-                    // Convert the string labels into key types.
-                    mlContext.Transforms.Conversion
-                        .MapValueToKey(nameof(DataPoint.Label))
-
-                    // Apply a multiclass trainer.
-                        .Append(mlContext.MulticlassClassification.Trainers
-                        .LightGbm());
+                // Convert the string labels into key types.
+                mlContext.Transforms.Conversion
+                .MapValueToKey(nameof(DataPoint.Label))
+                // Apply a multiclass trainer.
+                .Append(mlContext.MulticlassClassification.Trainers
+                .LightGbm());
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
index 7ccc793feb..47f5ea50cd 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
@@ -39,10 +39,10 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 <# if (MetaTrainer != null) { #>
             // Define the trainer.
             var pipeline =
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion.MapValueToKey("Label")
-            // Apply <#=MetaTrainer#> multiclass meta trainer on top of
-            // binary trainer.
+                // Apply <#=MetaTrainer#> multiclass meta trainer on top of
+                // binary trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .<#=MetaTrainer#>(
                 <#=Trainer#>()));
@@ -50,10 +50,10 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 <# } else if (TrainerOptions == null) { #>
             // Define the trainer.
             var pipeline =
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-            // Apply <#=Trainer#> multiclass trainer.
+                // Apply <#=Trainer#> multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .<#=Trainer#>());
 
@@ -63,9 +63,9 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
 
             // Define the trainer.
             var pipeline = 
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion.MapValueToKey("Label")
-            // Apply <#=Trainer#> multiclass trainer.
+                // Apply <#=Trainer#> multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .<#=Trainer#>(options));
             
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
index 33272b2fb0..d30413c663 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
@@ -31,10 +31,10 @@ public static void Example()
 
             // Define the trainer.
             var pipeline =
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-            // Apply NaiveBayes multiclass trainer.
+                // Apply NaiveBayes multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .NaiveBayes());
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
index 044cfcd3fb..e0dde48af7 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
@@ -25,10 +25,10 @@ public static void Example()
 
             // Define the trainer.
             var pipeline =
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion.MapValueToKey("Label")
-            // Apply OneVersusAll multiclass meta trainer on top of
-            // binary trainer.
+                // Apply OneVersusAll multiclass meta trainer on top of
+                // binary trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .OneVersusAll(
                 mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
index a2b518bc1a..2afb17e76c 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
@@ -25,10 +25,10 @@ public static void Example()
 
             // Define the trainer.
             var pipeline =
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion.MapValueToKey("Label")
-            // Apply PairwiseCoupling multiclass meta trainer on top of
-            // binary trainer.
+                // Apply PairwiseCoupling multiclass meta trainer on top of
+                // binary trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .PairwiseCoupling(
                 mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PermutationFeatureImportance.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PermutationFeatureImportance.cs
index 714c69361b..71d1c02106 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PermutationFeatureImportance.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PermutationFeatureImportance.cs
@@ -26,11 +26,11 @@ public static void Example()
                 new string[] { nameof(Data.Feature1), nameof(Data.Feature2) };
 
             var pipeline = mlContext.Transforms
-                    .Concatenate("Features", featureColumns)
-                    .Append(mlContext.Transforms.Conversion.MapValueToKey("Label"))
-                    .Append(mlContext.Transforms.NormalizeMinMax("Features"))
-                    .Append(mlContext.MulticlassClassification.Trainers
-                    .SdcaMaximumEntropy());
+                .Concatenate("Features", featureColumns)
+                .Append(mlContext.Transforms.Conversion.MapValueToKey("Label"))
+                .Append(mlContext.Transforms.NormalizeMinMax("Features"))
+                .Append(mlContext.MulticlassClassification.Trainers
+                .SdcaMaximumEntropy());
 
             // Fit the pipeline to the data.
             var model = pipeline.Fit(data);
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
index 6521c0e5e1..4b5608e30f 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
@@ -33,10 +33,10 @@ public static void Example()
 
             // Define the trainer.
             var pipeline =
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-            // Apply SdcaMaximumEntropy multiclass trainer.
+                // Apply SdcaMaximumEntropy multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaMaximumEntropy());
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
index b0a61c0061..4ec7c86cd9 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
@@ -43,9 +43,9 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion.MapValueToKey("Label")
-            // Apply SdcaMaximumEntropy multiclass trainer.
+                // Apply SdcaMaximumEntropy multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaMaximumEntropy(options));
             
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index 4e4a612428..dc3a2013c4 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -33,10 +33,10 @@ public static void Example()
 
             // Define the trainer.
             var pipeline =
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion
                 .MapValueToKey(nameof(DataPoint.Label))
-            // Apply SdcaNonCalibrated multiclass trainer.
+                // Apply SdcaNonCalibrated multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaNonCalibrated());
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
index d86bd7d1c5..f5004c3c84 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
@@ -43,9 +43,9 @@ public static void Example()
 
             // Define the trainer.
             var pipeline = 
-            // Convert the string labels into key types.
+                // Convert the string labels into key types.
                 mlContext.Transforms.Conversion.MapValueToKey("Label")
-            // Apply SdcaNonCalibrated multiclass trainer.
+                // Apply SdcaNonCalibrated multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaNonCalibrated(options));
             

From e497794faf0f193dc1dc815f8a442be77e9f5887 Mon Sep 17 00:00:00 2001
From: sierralee51 <36902786+sierralee51@users.noreply.github.com>
Date: Tue, 2 Jul 2019 14:33:35 -0700
Subject: [PATCH 43/44] fixed extra empty lines

---
 .../Trainers/MulticlassClassification/LightGbmWithOptions.cs   | 1 -
 .../MulticlassClassification.ttinclude                         | 3 ---
 .../Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs    | 1 -
 .../Trainers/MulticlassClassification/PairwiseCoupling.cs      | 1 -
 4 files changed, 6 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
index 4664fad79a..76337146dc 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbmWithOptions.cs
@@ -44,7 +44,6 @@ public static void Example()
                 // Apply LightGbm multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .LightGbm(options));
-            
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
index 47f5ea50cd..ee3f245566 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/MulticlassClassification.ttinclude
@@ -46,7 +46,6 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                 .Append(mlContext.MulticlassClassification.Trainers
                 .<#=MetaTrainer#>(
                 <#=Trainer#>()));
-
 <# } else if (TrainerOptions == null) { #>
             // Define the trainer.
             var pipeline =
@@ -56,7 +55,6 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                 // Apply <#=Trainer#> multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .<#=Trainer#>());
-
 <# } else { #>
             // Define trainer options.
             var options = new <#=TrainerOptions#>;
@@ -68,7 +66,6 @@ namespace Samples.Dynamic.Trainers.MulticlassClassification
                 // Apply <#=Trainer#> multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .<#=Trainer#>(options));
-            
 <# } #>
 
             // Train the model.
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
index d30413c663..4dc7f44f9e 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/NaiveBayes.cs
@@ -38,7 +38,6 @@ public static void Example()
                 .Append(mlContext.MulticlassClassification.Trainers
                 .NaiveBayes());
 
-
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
index 2afb17e76c..be42ec3bbd 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/PairwiseCoupling.cs
@@ -33,7 +33,6 @@ public static void Example()
                 .PairwiseCoupling(
                 mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
 
-
             // Train the model.
             var model = pipeline.Fit(trainingData);
 

From 6224ed5a76c8001208a29618d5cdba51631f2db7 Mon Sep 17 00:00:00 2001
From: sierralee51 <36902786+sierralee51@users.noreply.github.com>
Date: Tue, 2 Jul 2019 16:15:08 -0700
Subject: [PATCH 44/44] fixed some more indentation issue

---
 .../Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs     | 1 -
 .../MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs   | 1 -
 .../Dynamic/Trainers/MulticlassClassification/LightGbm.cs        | 1 -
 .../Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs    | 1 -
 .../Trainers/MulticlassClassification/SdcaMaximumEntropy.cs      | 1 -
 .../MulticlassClassification/SdcaMaximumEntropyWithOptions.cs    | 1 -
 .../Trainers/MulticlassClassification/SdcaNonCalibrated.cs       | 1 -
 .../MulticlassClassification/SdcaNonCalibratedWithOptions.cs     | 1 -
 8 files changed, 8 deletions(-)

diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
index 7670cf5f3d..7130a7c9ea 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropy.cs
@@ -32,7 +32,6 @@ public static void Example()
                 .Append(mlContext.MulticlassClassification.Trainers
                 .LbfgsMaximumEntropy());
 
-
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
index d99c7e0d6d..597f34af06 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LbfgsMaximumEntropyWithOptions.cs
@@ -39,7 +39,6 @@ public static void Example()
                 // Apply LbfgsMaximumEntropy multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .LbfgsMaximumEntropy(options));
-            
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
index 2cea68b970..7ab6d5f80e 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/LightGbm.cs
@@ -35,7 +35,6 @@ public static void Example()
                 .Append(mlContext.MulticlassClassification.Trainers
                 .LightGbm());
 
-
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
index e0dde48af7..40a040325c 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/OneVersusAll.cs
@@ -33,7 +33,6 @@ public static void Example()
                 .OneVersusAll(
                 mlContext.BinaryClassification.Trainers.SdcaLogisticRegression()));
 
-
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
index 4b5608e30f..2a83a84f3a 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropy.cs
@@ -40,7 +40,6 @@ public static void Example()
                 .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaMaximumEntropy());
 
-
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
index 4ec7c86cd9..62ed191c0f 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaMaximumEntropyWithOptions.cs
@@ -48,7 +48,6 @@ public static void Example()
                 // Apply SdcaMaximumEntropy multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaMaximumEntropy(options));
-            
 
             // Train the model.
             var model = pipeline.Fit(trainingData);
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
index dc3a2013c4..6a899f9432 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibrated.cs
@@ -40,7 +40,6 @@ public static void Example()
                 .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaNonCalibrated());
 
-
             // Train the model.
             var model = pipeline.Fit(trainingData);
 
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
index f5004c3c84..6367e9fe21 100644
--- a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
+++ b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/MulticlassClassification/SdcaNonCalibratedWithOptions.cs
@@ -48,7 +48,6 @@ public static void Example()
                 // Apply SdcaNonCalibrated multiclass trainer.
                 .Append(mlContext.MulticlassClassification.Trainers
                 .SdcaNonCalibrated(options));
-            
 
             // Train the model.
             var model = pipeline.Fit(trainingData);