StandardTrainersCatalog.AveragedPerceptron Method

Definition

Namespace:

Microsoft.ML

Assembly:

Microsoft.ML.StandardTrainers.dll

Package:

Microsoft.ML v3.0.1

Overloads

AveragedPerceptron(BinaryClassificationCatalog+BinaryClassificationTrainers, AveragedPerceptronTrainer+Options)	Create an AveragedPerceptronTrainer with advanced options, which predicts a target using a linear binary classification model trained over boolean label data.
AveragedPerceptron(BinaryClassificationCatalog+BinaryClassificationTrainers, String, String, IClassificationLoss, Single, Boolean, Single, Int32)	Create an AveragedPerceptronTrainer, which predicts a target using a linear binary classification model trained over boolean label data.

AveragedPerceptron(BinaryClassificationCatalog+BinaryClassificationTrainers, AveragedPerceptronTrainer+Options)

Create an AveragedPerceptronTrainer with advanced options, which predicts a target using a linear binary classification model trained over boolean label data.

public static Microsoft.ML.Trainers.AveragedPerceptronTrainer AveragedPerceptron (this Microsoft.ML.BinaryClassificationCatalog.BinaryClassificationTrainers catalog, Microsoft.ML.Trainers.AveragedPerceptronTrainer.Options options);

static member AveragedPerceptron : Microsoft.ML.BinaryClassificationCatalog.BinaryClassificationTrainers * Microsoft.ML.Trainers.AveragedPerceptronTrainer.Options -> Microsoft.ML.Trainers.AveragedPerceptronTrainer

<Extension()>
Public Function AveragedPerceptron (catalog As BinaryClassificationCatalog.BinaryClassificationTrainers, options As AveragedPerceptronTrainer.Options) As AveragedPerceptronTrainer

Parameters

catalog

BinaryClassificationCatalog.BinaryClassificationTrainers

The binary classification catalog trainer object.

options

AveragedPerceptronTrainer.Options

Trainer options.

Returns

AveragedPerceptronTrainer

Examples

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 AveragedPerceptronWithOptions
    {
        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 AveragedPerceptronTrainer.Options
            {
                LossFunction = new SmoothedHingeLoss(),
                LearningRate = 0.1f,
                LazyUpdate = false,
                RecencyGain = 0.1f,
                NumberOfIterations = 10
            };

            // Define the trainer.
            var pipeline = mlContext.BinaryClassification.Trainers
                .AveragedPerceptron(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();

            // Print 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: True
            //   Label: False, Prediction: False

            // Evaluate the overall metrics.
            var metrics = mlContext.BinaryClassification
                .EvaluateNonCalibrated(transformedTestData);

            PrintMetrics(metrics);

            // Expected output:
            //   Accuracy: 0.89
            //   AUC: 0.96
            //   F1 Score: 0.88
            //   Negative Precision: 0.87
            //   Negative Recall: 0.92
            //   Positive Precision: 0.91
            //   Positive Recall: 0.85
            //
            // TEST POSITIVE RATIO:    0.4760 (238.0/(238.0+262.0))
            //   Confusion table
            //             ||======================
            //   PREDICTED || positive | negative | Recall
            //   TRUTH     ||======================
            //    positive ||      151 |       87 | 0.6345
            //    negative ||       53 |      209 | 0.7977
            //             ||======================
            //   Precision ||   0.7402 |   0.7061 |
        }

        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() +
                        0.1f).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; }
        }

        // Pretty-print BinaryClassificationMetrics objects.
        private static void PrintMetrics(BinaryClassificationMetrics metrics)
        {
            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}\n");
            Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
        }
    }
}

AveragedPerceptron(BinaryClassificationCatalog+BinaryClassificationTrainers, String, String, IClassificationLoss, Single, Boolean, Single, Int32)

Create an AveragedPerceptronTrainer, which predicts a target using a linear binary classification model trained over boolean label data.

public static Microsoft.ML.Trainers.AveragedPerceptronTrainer AveragedPerceptron (this Microsoft.ML.BinaryClassificationCatalog.BinaryClassificationTrainers catalog, string labelColumnName = "Label", string featureColumnName = "Features", Microsoft.ML.Trainers.IClassificationLoss lossFunction = default, float learningRate = 1, bool decreaseLearningRate = false, float l2Regularization = 0, int numberOfIterations = 10);

static member AveragedPerceptron : Microsoft.ML.BinaryClassificationCatalog.BinaryClassificationTrainers * string * string * Microsoft.ML.Trainers.IClassificationLoss * single * bool * single * int -> Microsoft.ML.Trainers.AveragedPerceptronTrainer

<Extension()>
Public Function AveragedPerceptron (catalog As BinaryClassificationCatalog.BinaryClassificationTrainers, Optional labelColumnName As String = "Label", Optional featureColumnName As String = "Features", Optional lossFunction As IClassificationLoss = Nothing, Optional learningRate As Single = 1, Optional decreaseLearningRate As Boolean = false, Optional l2Regularization As Single = 0, Optional numberOfIterations As Integer = 10) As AveragedPerceptronTrainer

Parameters

catalog

BinaryClassificationCatalog.BinaryClassificationTrainers

The binary classification catalog trainer object.

labelColumnName

String

The name of the label column. The column data must be Boolean.

featureColumnName

String

The name of the feature column. The column data must be a known-sized vector of Single.

lossFunction

IClassificationLoss

The loss function minimized in the training process. If null, HingeLoss would be used and lead to a max-margin averaged perceptron trainer.

learningRate

Single

The initial learning rate used by SGD.

decreaseLearningRate

Boolean

true to decrease the learningRate as iterations progress; otherwise, false. Default is false.

l2Regularization

Single

The L2 weight for regularization.

numberOfIterations

Int32

Number of passes through the training dataset.

Returns

AveragedPerceptronTrainer

Examples

using System;
using System.Collections.Generic;
using System.Linq;
using Microsoft.ML;
using Microsoft.ML.Data;

namespace Samples.Dynamic.Trainers.BinaryClassification
{
    public static class AveragedPerceptron
    {
        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
                .AveragedPerceptron();

            // 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();

            // Print 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: False

            // Evaluate the overall metrics.
            var metrics = mlContext.BinaryClassification
                .EvaluateNonCalibrated(transformedTestData);

            PrintMetrics(metrics);

            // Expected output:
            //   Accuracy: 0.72
            //   AUC: 0.79
            //   F1 Score: 0.68
            //   Negative Precision: 0.71
            //   Negative Recall: 0.80
            //   Positive Precision: 0.74
            //   Positive Recall: 0.63
            //
            //   TEST POSITIVE RATIO:    0.4760 (238.0/(238.0+262.0))
            //   Confusion table
            //             ||======================
            //   PREDICTED || positive | negative | Recall
            //   TRUTH     ||======================
            //    positive ||      151 |       87 | 0.6345
            //    negative ||       53 |      209 | 0.7977
            //             ||======================
            //   Precision ||   0.7402 |   0.7061 |
        }

        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() +
                        0.1f).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; }
        }

        // Pretty-print BinaryClassificationMetrics objects.
        private static void PrintMetrics(BinaryClassificationMetrics metrics)
        {
            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}\n");
            Console.WriteLine(metrics.ConfusionMatrix.GetFormattedConfusionTable());
        }
    }
}