linear svm

dotnet · Apr 12, 2019 · ace5d1b · ace5d1b
1 parent 273eba2
commit ace5d1b
Show file tree

Hide file tree

Showing 6 changed files with 300 additions and 27 deletions.
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/LinearSvm.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/LinearSvm.cs
@@ -0,0 +1,104 @@
+using System;
+using System.Collections.Generic;
+using System.Linq;
+using Microsoft.ML;
+using Microsoft.ML.Data;
+
+namespace Samples.Dynamic.Trainers.BinaryClassification
+{
+    public static class LinearSvm
+    {
+        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.
+            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.
+            var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
+
+            // Define the trainer.
+            var pipeline = mlContext.BinaryClassification.Trainers.LinearSvm();
+
+            // 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));
+
+            // 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();
+
+            // Look at 5 predictions
+            foreach (var p in predictions.Take(5))
+                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+
+            // Expected output:
+            //   Label: True, Prediction: True
+            //   Label: False, Prediction: False
+            //   Label: True, Prediction: True
+            //   Label: True, Prediction: False
+            //   Label: False, Prediction: True
+
+            // Evaluate the overall metrics
+            var metrics = mlContext.BinaryClassification.EvaluateNonCalibrated(transformedTestData);
+            Console.WriteLine($"Accuracy: {metrics.Accuracy:F2}");
+            Console.WriteLine($"AUC: {metrics.AreaUnderRocCurve:F2}");
+            Console.WriteLine($"F1 Score: {metrics.F1Score:F2}");
+            Console.WriteLine($"Negative Precision: {metrics.NegativePrecision:F2}");
+            Console.WriteLine($"Negative Recall: {metrics.NegativeRecall:F2}");
+            Console.WriteLine($"Positive Precision: {metrics.PositivePrecision:F2}");
+            Console.WriteLine($"Positive Recall: {metrics.PositiveRecall:F2}");
+
+            // Expected output:
+            //   Accuracy: 0.53
+            //   AUC: 0.56
+            //   F1 Score: 0.49
+            //   Negative Precision: 0.55
+            //   Negative Recall: 0.58
+            //   Positive Precision: 0.50
+            //   Positive Recall: 0.47
+        }
+
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        {
+            var random = new Random(seed);
+            float randomFloat() => (float)random.NextDouble();
+            for (int i = 0; i < count; i++)
+            {
+                var label = randomFloat() > 0.5f;
+                yield return new DataPoint
+                {
+                    Label = label,
+                    // Create random features that are correlated with the label.
+                    // For data points with false label, the feature values are slightly increased by adding a constant.
+                    Features = Enumerable.Repeat(label, 50).Select(x => x ? randomFloat() : randomFloat() + 0.03f).ToArray()
+                };
+            }
+        }
+
+        // Example with label and 50 feature values. A data set is a collection of such examples.
+        private class DataPoint
+        {
+            public bool Label { get; set; }
+            [VectorType(50)]
+            public float[] Features { get; set; }
+        }
+
+        // Class used to capture predictions.
+        private class Prediction
+        {
+            // Original label.
+            public bool Label { get; set; }
+            // Predicted label from the trainer.
+            public bool PredictedLabel { get; set; }
+        }
+    }
+}
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/LinearSvm.tt b/docs/samples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/LinearSvm.tt
@@ -0,0 +1,28 @@
+<#@ include file="BinaryClassification.ttinclude"#>
+<#+
+string ClassName = "LinearSvm";
+string Trainer = "LinearSvm";
+string TrainerOptions = null;
+bool IsCalibrated = false;
+bool CacheData = false;
+
+string DataSepValue = "0.03f";
+string OptionsInclude = "";
+string Comments = "";
+
+string ExpectedOutputPerInstance = @"// Expected output:
+            //   Label: True, Prediction: True
+            //   Label: False, Prediction: False
+            //   Label: True, Prediction: True
+            //   Label: True, Prediction: False
+            //   Label: False, Prediction: True";
+
+string ExpectedOutput = @"// Expected output:
+            //   Accuracy: 0.53
+            //   AUC: 0.56
+            //   F1 Score: 0.49
+            //   Negative Precision: 0.55
+            //   Negative Recall: 0.58
+            //   Positive Precision: 0.50
+            //   Positive Recall: 0.47";
+#>
diff --git a/...amples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/LinearSvmWithOptions.cs b/...amples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/LinearSvmWithOptions.cs
@@ -0,0 +1,113 @@
+using System;
+using System.Collections.Generic;
+using System.Linq;
+using Microsoft.ML;
+using Microsoft.ML.Data;
+using Microsoft.ML.Trainers;
+
+namespace Samples.Dynamic.Trainers.BinaryClassification
+{
+    public static class LinearSvmWithOptions
+    {
+        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.
+            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.
+            var trainingData = mlContext.Data.LoadFromEnumerable(dataPoints);
+
+            // Define trainer options.
+            var options = new LinearSvmTrainer.Options
+            {
+                BatchSize = 10,
+                PerformProjection = true,
+                NumberOfIterations = 10
+            };
+
+            // Define the trainer.
+            var pipeline = mlContext.BinaryClassification.Trainers.LinearSvm(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));
+
+            // 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();
+
+            // Look at 5 predictions
+            foreach (var p in predictions.Take(5))
+                Console.WriteLine($"Label: {p.Label}, Prediction: {p.PredictedLabel}");
+
+            // Expected output:
+            //   Label: True, Prediction: False
+            //   Label: False, Prediction: False
+            //   Label: True, Prediction: True
+            //   Label: True, Prediction: True
+            //   Label: False, Prediction: False
+
+            // Evaluate the overall metrics
+            var metrics = mlContext.BinaryClassification.EvaluateNonCalibrated(transformedTestData);
+            Console.WriteLine($"Accuracy: {metrics.Accuracy:F2}");
+            Console.WriteLine($"AUC: {metrics.AreaUnderRocCurve:F2}");
+            Console.WriteLine($"F1 Score: {metrics.F1Score:F2}");
+            Console.WriteLine($"Negative Precision: {metrics.NegativePrecision:F2}");
+            Console.WriteLine($"Negative Recall: {metrics.NegativeRecall:F2}");
+            Console.WriteLine($"Positive Precision: {metrics.PositivePrecision:F2}");
+            Console.WriteLine($"Positive Recall: {metrics.PositiveRecall:F2}");
+
+            // Expected output:
+            //   Accuracy: 0.58
+            //   AUC: 0.67
+            //   F1 Score: 0.28
+            //   Negative Precision: 0.56
+            //   Negative Recall: 0.94
+            //   Positive Precision: 0.74
+            //   Positive Recall: 0.18
+        }
+
+        private static IEnumerable<DataPoint> GenerateRandomDataPoints(int count, int seed=0)
+        {
+            var random = new Random(seed);
+            float randomFloat() => (float)random.NextDouble();
+            for (int i = 0; i < count; i++)
+            {
+                var label = randomFloat() > 0.5f;
+                yield return new DataPoint
+                {
+                    Label = label,
+                    // Create random features that are correlated with the label.
+                    // For data points with false label, the feature values are slightly increased by adding a constant.
+                    Features = Enumerable.Repeat(label, 50).Select(x => x ? randomFloat() : randomFloat() + 0.03f).ToArray()
+                };
+            }
+        }
+
+        // Example with label and 50 feature values. A data set is a collection of such examples.
+        private class DataPoint
+        {
+            public bool Label { get; set; }
+            [VectorType(50)]
+            public float[] Features { get; set; }
+        }
+
+        // Class used to capture predictions.
+        private class Prediction
+        {
+            // Original label.
+            public bool Label { get; set; }
+            // Predicted label from the trainer.
+            public bool PredictedLabel { get; set; }
+        }
+    }
+}
diff --git a/...amples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/LinearSvmWithOptions.tt b/...amples/Microsoft.ML.Samples/Dynamic/Trainers/BinaryClassification/LinearSvmWithOptions.tt
@@ -0,0 +1,34 @@
+<#@ include file="BinaryClassification.ttinclude"#>
+<#+
+string ClassName = "LinearSvmWithOptions";
+string Trainer = "LinearSvm";
+bool IsCalibrated = false;
+
+string DataSepValue = "0.03f";
+string OptionsInclude = "using Microsoft.ML.Trainers;";
+string Comments = "";
+bool CacheData = false;
+
+string TrainerOptions = @"LinearSvmTrainer.Options
+            {
+                BatchSize = 10,
+                PerformProjection = true,
+                NumberOfIterations = 10
+            }";
+
+string ExpectedOutputPerInstance = @"// Expected output:
+            //   Label: True, Prediction: False
+            //   Label: False, Prediction: False
+            //   Label: True, Prediction: True
+            //   Label: True, Prediction: True
+            //   Label: False, Prediction: False";
+
+string ExpectedOutput = @"// Expected output:
+            //   Accuracy: 0.58
+            //   AUC: 0.67
+            //   F1 Score: 0.28
+            //   Negative Precision: 0.56
+            //   Negative Recall: 0.94
+            //   Positive Precision: 0.74
+            //   Positive Recall: 0.18";
+#>
diff --git a/docs/samples/Microsoft.ML.Samples/Microsoft.ML.Samples.csproj b/docs/samples/Microsoft.ML.Samples/Microsoft.ML.Samples.csproj
@@ -108,6 +108,14 @@
       <Generator>TextTemplatingFileGenerator</Generator>
       <LastGenOutput>LightGbmWithOptions.cs</LastGenOutput>
     </None>
+    <None Update="Dynamic\Trainers\BinaryClassification\LinearSvm.tt">
+      <Generator>TextTemplatingFileGenerator</Generator>
+      <LastGenOutput>LinearSvm.cs</LastGenOutput>
+    </None>
+    <None Update="Dynamic\Trainers\BinaryClassification\LinearSvmWithOptions.tt">
+      <Generator>TextTemplatingFileGenerator</Generator>
+      <LastGenOutput>LinearSvmWithOptions.cs</LastGenOutput>
+    </None>
     <None Update="Dynamic\Trainers\BinaryClassification\SdcaLogisticRegression.tt">
       <Generator>TextTemplatingFileGenerator</Generator>
       <LastGenOutput>SdcaLogisticRegression.cs</LastGenOutput>
@@ -229,6 +237,16 @@
       <AutoGen>True</AutoGen>
       <DependentUpon>LightGbmWithOptions.tt</DependentUpon>
     </Compile>
+    <Compile Update="Dynamic\Trainers\BinaryClassification\LinearSvm.cs">
+      <DesignTime>True</DesignTime>
+      <AutoGen>True</AutoGen>
+      <DependentUpon>LinearSvm.tt</DependentUpon>
+    </Compile>
+    <Compile Update="Dynamic\Trainers\BinaryClassification\LinearSvmWithOptions.cs">
+      <DesignTime>True</DesignTime>
+      <AutoGen>True</AutoGen>
+      <DependentUpon>LinearSvmWithOptions.tt</DependentUpon>
+    </Compile>
     <Compile Update="Dynamic\Trainers\BinaryClassification\PriorTrainer.cs">
       <DesignTime>True</DesignTime>
       <AutoGen>True</AutoGen>

diff --git a/docs/samples/Microsoft.ML.Samples/Program.cs b/docs/samples/Microsoft.ML.Samples/Program.cs
@@ -1,36 +1,12 @@
-using System;
-using Samples.Dynamic.Trainers.BinaryClassification;
+using Samples.Dynamic;
 
 namespace Microsoft.ML.Samples
 {
     internal static class Program
     {
         static void Main(string[] args)
         {
-            Console.WriteLine("Prior");
-            Prior.Example();
-            Console.WriteLine("\nSdcaLogisticRegression");
-            SdcaLogisticRegression.Example();
-            Console.WriteLine("\nSdcaLogisticRegressionWithOptions");
-            SdcaLogisticRegressionWithOptions.Example();
-            Console.WriteLine("\nSdcaNonCalibrated");
-            SdcaNonCalibrated.Example();
-            Console.WriteLine("\nSdcaNonCalibratedWithOptions");
-            SdcaNonCalibratedWithOptions.Example();
-            Console.WriteLine("\nSgdCalibrated");
-            SgdCalibrated.Example();
-            Console.WriteLine("\nSgdCalibratedWithOptions");
-            SgdCalibratedWithOptions.Example();
-            Console.WriteLine("\nSgdNonCalibrated");
-            SgdNonCalibrated.Example();
-            Console.WriteLine("\nSgdNonCalibratedWithOptions");
-            SgdNonCalibratedWithOptions.Example();
-            Console.WriteLine("\nSymbolicSgdLogisticRegression");
-            SymbolicSgdLogisticRegression.Example();
-            Console.WriteLine("\nSymbolicSgdLogisticRegressionWithOptions");
-            SymbolicSgdLogisticRegression.Example();
-
-            Console.ReadLine();
+            CalculateFeatureContribution.Example();
         }
     }
-}
+}