TimeSeriesCatalog.DetectSpikeBySsa 方法

参考

定义

命名空间:: Microsoft.ML

程序集:: Microsoft.ML.TimeSeries.dll

包:: Microsoft.ML.TimeSeries v3.0.1

包:: Microsoft.ML.TimeSeries v1.5.5

包:: Microsoft.ML.TimeSeries v1.6.0

包:: Microsoft.ML.TimeSeries v1.7.0

包:: Microsoft.ML.TimeSeries v2.0.0

包:: Microsoft.ML.TimeSeries v1.2.0

包:: Microsoft.ML.TimeSeries v1.3.1

包:: Microsoft.ML.TimeSeries v1.4.0

重要

一些信息与预发行产品相关，相应产品在发行之前可能会进行重大修改。对于此处提供的信息，Microsoft 不作任何明示或暗示的担保。

重载

DetectSpikeBySsa(TransformsCatalog, String, String, Double, Int32, Int32, Int32, AnomalySide, ErrorFunction)	创建 SsaSpikeEstimator，它使用单数光谱分析 (SSA) 预测时序峰值。
DetectSpikeBySsa(TransformsCatalog, String, String, Int32, Int32, Int32, Int32, AnomalySide, ErrorFunction)	已过时. 创建 SsaSpikeEstimator，它使用单数光谱分析 (SSA) 预测时序峰值。

DetectSpikeBySsa(TransformsCatalog, String, String, Double, Int32, Int32, Int32, AnomalySide, ErrorFunction)

创建 SsaSpikeEstimator，它使用单数光谱分析 (SSA) 预测时序峰值。

public static Microsoft.ML.Transforms.TimeSeries.SsaSpikeEstimator DetectSpikeBySsa (this Microsoft.ML.TransformsCatalog catalog, string outputColumnName, string inputColumnName, double confidence, int pvalueHistoryLength, int trainingWindowSize, int seasonalityWindowSize, Microsoft.ML.Transforms.TimeSeries.AnomalySide side = Microsoft.ML.Transforms.TimeSeries.AnomalySide.TwoSided, Microsoft.ML.Transforms.TimeSeries.ErrorFunction errorFunction = Microsoft.ML.Transforms.TimeSeries.ErrorFunction.SignedDifference);

static member DetectSpikeBySsa : Microsoft.ML.TransformsCatalog * string * string * double * int * int * int * Microsoft.ML.Transforms.TimeSeries.AnomalySide * Microsoft.ML.Transforms.TimeSeries.ErrorFunction -> Microsoft.ML.Transforms.TimeSeries.SsaSpikeEstimator

<Extension()>
Public Function DetectSpikeBySsa (catalog As TransformsCatalog, outputColumnName As String, inputColumnName As String, confidence As Double, pvalueHistoryLength As Integer, trainingWindowSize As Integer, seasonalityWindowSize As Integer, Optional side As AnomalySide = Microsoft.ML.Transforms.TimeSeries.AnomalySide.TwoSided, Optional errorFunction As ErrorFunction = Microsoft.ML.Transforms.TimeSeries.ErrorFunction.SignedDifference) As SsaSpikeEstimator

参数

catalog: TransformsCatalog

转换的目录。

outputColumnName: String

由转换 inputColumnName生成的列的名称。列数据是一个向量 Double。矢量包含 3 个元素：警报 (非零值表示峰值) 、原始分数和 p 值。

inputColumnName: String

要转换的列的名称。列数据必须是 Single。 If set to null, the value of the outputColumnName will be used as source.

confidence: Double

[0， 100] 范围内的峰值检测置信度。

pvalueHistoryLength: Int32

用于计算 p 值的滑动窗口的大小。

trainingWindowSize: Int32

用于训练的序列开头的点数。

seasonalityWindowSize: Int32

输入时序中最大相关季节性的上限。

side: AnomalySide

确定是检测正异常还是负异常还是同时检测两者的参数。

errorFunction: ErrorFunction

用于计算预期值和观察到值之间的错误的函数。

SsaSpikeEstimator

示例

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

namespace Samples.Dynamic
{
    public static class DetectSpikeBySsaBatchPrediction
    {
        // This example creates a time series (list of Data with the i-th element
        // corresponding to the i-th time slot). The estimator is applied then to
        // identify spiking points in the series. This estimator can account for
        // temporal seasonality in the data.
        public static void Example()
        {
            // Create a new ML context, for ML.NET operations. It can be used for
            // exception tracking and logging, as well as the source of randomness.
            var ml = new MLContext();

            // Generate sample series data with a recurring pattern and a spike
            // within the pattern
            const int SeasonalitySize = 5;
            const int TrainingSeasons = 3;
            const int TrainingSize = SeasonalitySize * TrainingSeasons;
            var data = new List<TimeSeriesData>()
            {
                new TimeSeriesData(0),
                new TimeSeriesData(1),
                new TimeSeriesData(2),
                new TimeSeriesData(3),
                new TimeSeriesData(4),

                new TimeSeriesData(0),
                new TimeSeriesData(1),
                new TimeSeriesData(2),
                new TimeSeriesData(3),
                new TimeSeriesData(4),

                new TimeSeriesData(0),
                new TimeSeriesData(1),
                new TimeSeriesData(2),
                new TimeSeriesData(3),
                new TimeSeriesData(4),

                //This is a spike.
                new TimeSeriesData(100),

                new TimeSeriesData(0),
                new TimeSeriesData(1),
                new TimeSeriesData(2),
                new TimeSeriesData(3),
                new TimeSeriesData(4),
            };

            // Convert data to IDataView.
            var dataView = ml.Data.LoadFromEnumerable(data);

            // Setup estimator arguments
            var inputColumnName = nameof(TimeSeriesData.Value);
            var outputColumnName = nameof(SsaSpikePrediction.Prediction);

            // The transformed data.
            var transformedData = ml.Transforms.DetectSpikeBySsa(outputColumnName,
                inputColumnName, 95.0d, 8, TrainingSize, SeasonalitySize + 1).Fit(
                dataView).Transform(dataView);

            // Getting the data of the newly created column as an IEnumerable of
            // SsaSpikePrediction.
            var predictionColumn = ml.Data.CreateEnumerable<SsaSpikePrediction>(
                transformedData, reuseRowObject: false);

            Console.WriteLine($"{outputColumnName} column obtained " +
                $"post-transformation.");

            Console.WriteLine("Data\tAlert\tScore\tP-Value");
            int k = 0;
            foreach (var prediction in predictionColumn)
                PrintPrediction(data[k++].Value, prediction);

            // Prediction column obtained post-transformation.
            // Data    Alert   Score   P-Value
            // 0       0      -2.53    0.50
            // 1       0      -0.01    0.01
            // 2       0       0.76    0.14
            // 3       0       0.69    0.28
            // 4       0       1.44    0.18
            // 0       0      -1.84    0.17
            // 1       0       0.22    0.44
            // 2       0       0.20    0.45
            // 3       0       0.16    0.47
            // 4       0       1.33    0.18
            // 0       0      -1.79    0.07
            // 1       0       0.16    0.50
            // 2       0       0.09    0.50
            // 3       0       0.08    0.45
            // 4       0       1.31    0.12
            // 100     1      98.21    0.00   <-- alert is on, predicted spike
            // 0       0     -13.83    0.29
            // 1       0      -1.74    0.44
            // 2       0      -0.47    0.46
            // 3       0     -16.50    0.29
            // 4       0     -29.82    0.21
        }

        private static void PrintPrediction(float value, SsaSpikePrediction
            prediction) =>
            Console.WriteLine("{0}\t{1}\t{2:0.00}\t{3:0.00}", value,
            prediction.Prediction[0], prediction.Prediction[1],
            prediction.Prediction[2]);

        class TimeSeriesData
        {
            public float Value;

            public TimeSeriesData(float value)
            {
                Value = value;
            }
        }

        class SsaSpikePrediction
        {
            [VectorType(3)]
            public double[] Prediction { get; set; }
        }
    }
}

适用于

DetectSpikeBySsa(TransformsCatalog, String, String, Int32, Int32, Int32, Int32, AnomalySide, ErrorFunction)

注意

This API method is deprecated, please use the overload with confidence parameter of type double.