PermutationFeatureImportanceExtensions.PermutationFeatureImportance Metoda
Definice
Důležité
Některé informace platí pro předběžně vydaný produkt, který se může zásadně změnit, než ho výrobce nebo autor vydá. Microsoft neposkytuje žádné záruky, výslovné ani předpokládané, týkající se zde uváděných informací.
Přetížení
PermutationFeatureImportance(MulticlassClassificationCatalog, ITransformer, IDataView, String, Boolean, Nullable<Int32>, Int32)
Důležitost funkce permutace (PFI) pro multiclassClassification
public static System.Collections.Immutable.ImmutableDictionary<string,Microsoft.ML.Data.MulticlassClassificationMetricsStatistics> PermutationFeatureImportance (this Microsoft.ML.MulticlassClassificationCatalog catalog, Microsoft.ML.ITransformer model, Microsoft.ML.IDataView data, string labelColumnName = "Label", bool useFeatureWeightFilter = false, int? numberOfExamplesToUse = default, int permutationCount = 1);
static member PermutationFeatureImportance : Microsoft.ML.MulticlassClassificationCatalog * Microsoft.ML.ITransformer * Microsoft.ML.IDataView * string * bool * Nullable<int> * int -> System.Collections.Immutable.ImmutableDictionary<string, Microsoft.ML.Data.MulticlassClassificationMetricsStatistics>
<Extension()>
Public Function PermutationFeatureImportance (catalog As MulticlassClassificationCatalog, model As ITransformer, data As IDataView, Optional labelColumnName As String = "Label", Optional useFeatureWeightFilter As Boolean = false, Optional numberOfExamplesToUse As Nullable(Of Integer) = Nothing, Optional permutationCount As Integer = 1) As ImmutableDictionary(Of String, MulticlassClassificationMetricsStatistics)
Parametry
- catalog
- MulticlassClassificationCatalog
Katalog klasifikace s více třídami.
- model
- ITransformer
Model, na kterém se má vyhodnotit důležitost funkce.
- data
- IDataView
Sada dat vyhodnocení.
- labelColumnName
- String
Název sloupce popisku Data sloupce musí být KeyDataViewType.
- useFeatureWeightFilter
- Boolean
Pomocí váhy funkcí můžete předfiltrovat funkce.
Omezte počet příkladů, na které se mají vyhodnotit. znamená, že se použijí až ~2 bln příklady.
- permutationCount
- Int32
Počet permutací, které se mají provést.
Návraty
Slovník mapuje každou funkci na každou funkci "příspěvky" na skóre.
Příklady
using System;
using System.Collections.Generic;
using System.Linq;
using Microsoft.ML;
namespace Samples.Dynamic.Trainers.MulticlassClassification
{
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 sample data.
var samples = GenerateData();
// 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)
.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);
// Transform the dataset.
var transformedData = model.Transform(data);
// 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 })
.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();
foreach (int i in sortedIndices)
{
Console.WriteLine("{0}\t{1:G4}\t{2:G4}",
featureColumns[i],
microAccuracy[i].Mean,
1.96 * microAccuracy[i].StandardError);
}
// Expected output:
//Feature Change in MicroAccuracy 95% Confidence in the Mean Change in MicroAccuracy
//Feature2 -0.1395 0.0006567
//Feature1 -0.05367 0.0006908
}
private class Data
{
public float Label { get; set; }
public float Feature1 { get; set; }
public float Feature2 { get; set; }
}
/// <summary>
/// Generate an enumerable of Data objects, creating the label as a simple
/// 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>
/// <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)
{
var rng = new Random(seed);
var max = bias + 4.5 * weight1 + 4.5 * weight2 + 0.5;
for (int i = 0; i < nExamples; i++)
{
var data = new Data
{
Feature1 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
Feature2 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
};
// Create a noisy label.
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)
data.Label = 1;
else
data.Label = 2;
yield return data;
}
}
}
}
Poznámky
Důležitost funkce permutace (PFI) je technika k určení globální důležitosti funkcí v natrénovaného modelu strojového učení. PFI je jednoduchá, ale výkonná technika motivovaná Breimanem ve svém dokumentu Random Forest, oddíl 10 (Breiman. "Náhodné doménové struktury". Machine Learning, 2001.) Výhodou metody PFI je, že je nezávislá na modelu – funguje s jakýmkoli modelem, který lze vyhodnotit – a může používat libovolnou datovou sadu, nejen trénovací sadu, k výpočtu metrik důležitosti funkcí.
PFI funguje tak, že vezme datovou sadu označenou jako datovou sadu, zvolí funkci a probere hodnoty této funkce napříč všemi příklady, takže každý příklad teď má pro tuto funkci náhodnou hodnotu a původní hodnoty pro všechny ostatní funkce. Metrika vyhodnocení (např. mikropřesnost) se pak vypočítá pro tuto upravenou datovou sadu a změna vyhodnocovací metriky z původní datové sady se vypočítá. Čím větší je změna metriky vyhodnocení, tím důležitější je funkce modelu. PFI funguje provedením této analýzy permutace napříč všemi funkcemi modelu, jeden za druhým.
V této implementaci PFI vypočítá změnu ve všech možných metrikách vyhodnocení klasifikace více tříd pro každou funkci a ImmutableArray vrátí se objekty MulticlassClassificationMetrics . Příklad práce s těmito výsledky pro analýzu důležitosti funkce modelu najdete v následující ukázce.
Platí pro
PermutationFeatureImportance(RegressionCatalog, ITransformer, IDataView, String, Boolean, Nullable<Int32>, Int32)
Důležitost funkce permutace (PFI) pro regresi
public static System.Collections.Immutable.ImmutableDictionary<string,Microsoft.ML.Data.RegressionMetricsStatistics> PermutationFeatureImportance (this Microsoft.ML.RegressionCatalog catalog, Microsoft.ML.ITransformer model, Microsoft.ML.IDataView data, string labelColumnName = "Label", bool useFeatureWeightFilter = false, int? numberOfExamplesToUse = default, int permutationCount = 1);
static member PermutationFeatureImportance : Microsoft.ML.RegressionCatalog * Microsoft.ML.ITransformer * Microsoft.ML.IDataView * string * bool * Nullable<int> * int -> System.Collections.Immutable.ImmutableDictionary<string, Microsoft.ML.Data.RegressionMetricsStatistics>
<Extension()>
Public Function PermutationFeatureImportance (catalog As RegressionCatalog, model As ITransformer, data As IDataView, Optional labelColumnName As String = "Label", Optional useFeatureWeightFilter As Boolean = false, Optional numberOfExamplesToUse As Nullable(Of Integer) = Nothing, Optional permutationCount As Integer = 1) As ImmutableDictionary(Of String, RegressionMetricsStatistics)
Parametry
- catalog
- RegressionCatalog
Regresní katalog.
- model
- ITransformer
Model, na kterém se má vyhodnotit důležitost funkce.
- data
- IDataView
Sada dat vyhodnocení.
- useFeatureWeightFilter
- Boolean
Pomocí váhy funkcí můžete předfiltrovat funkce.
Omezte počet příkladů, na které se mají vyhodnotit. znamená, že se použijí až ~2 bln příklady.
- permutationCount
- Int32
Počet permutací, které se mají provést.
Návraty
Slovník mapuje každou funkci na každou funkci "příspěvky" na skóre.
Příklady
using System;
using System.Collections.Generic;
using System.Linq;
using Microsoft.ML;
namespace Samples.Dynamic.Trainers.Regression
{
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 sample data.
var samples = GenerateData();
// 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)
.Append(mlContext.Transforms.NormalizeMinMax("Features"))
.Append(mlContext.Regression.Trainers.Ols());
// Fit the pipeline to the data.
var model = pipeline.Fit(data);
// Transform the dataset.
var transformedData = model.Transform(data);
// Extract the predictor.
var linearPredictor = model.LastTransformer;
// Compute the permutation metrics for the linear model using the
// normalized data.
var permutationMetrics = mlContext.Regression
.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 RMSE.
var sortedIndices = permutationMetrics
.Select((metrics, index) => new
{
index,
metrics.RootMeanSquaredError
})
.OrderByDescending(feature => Math.Abs(
feature.RootMeanSquaredError.Mean))
.Select(feature => feature.index);
Console.WriteLine("Feature\tModel Weight\tChange in RMSE\t95%" +
"Confidence in the Mean Change in RMSE");
var rmse = permutationMetrics.Select(x => x.RootMeanSquaredError)
.ToArray();
foreach (int i in sortedIndices)
{
Console.WriteLine("{0}\t{1:0.00}\t{2:G4}\t{3:G4}",
featureColumns[i],
linearPredictor.Model.Weights[i],
rmse[i].Mean,
1.96 * rmse[i].StandardError);
}
// Expected output:
// Feature Model Weight Change in RMSE 95% Confidence in the Mean Change in RMSE
// Feature2 9.00 4.009 0.008304
// Feature1 4.48 1.901 0.003351
}
private class Data
{
public float Label { get; set; }
public float Feature1 { get; set; }
public float Feature2 { get; set; }
}
/// <summary>
/// Generate an enumerable of Data objects, creating the label as a simple
/// 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>
/// <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)
{
var rng = new Random(seed);
for (int i = 0; i < nExamples; i++)
{
var data = new Data
{
Feature1 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
Feature2 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
};
// Create a noisy label.
data.Label = (float)(bias + weight1 * data.Feature1 + weight2 *
data.Feature2 + rng.NextDouble() - 0.5);
yield return data;
}
}
}
}
Poznámky
Důležitost funkce permutace (PFI) je technika k určení globální důležitosti funkcí v natrénovaného modelu strojového učení. PFI je jednoduchá, ale výkonná technika motivovaná Breimanem ve svém dokumentu Random Forest, oddíl 10 (Breiman. "Náhodné doménové struktury". Machine Learning, 2001.) Výhodou metody PFI je, že je nezávislá na modelu – funguje s jakýmkoli modelem, který lze vyhodnotit – a může používat libovolnou datovou sadu, nejen trénovací sadu, k výpočtu metrik důležitosti funkcí.
PFI funguje tak, že vezme datovou sadu označenou jako datovou sadu, zvolí funkci a probere hodnoty této funkce napříč všemi příklady, takže každý příklad teď má pro tuto funkci náhodnou hodnotu a původní hodnoty pro všechny ostatní funkce. Metrika vyhodnocení (např. R-squared) se pak vypočítá pro tuto upravenou datovou sadu a změna vyhodnocovací metriky z původní datové sady se vypočítá. Čím větší je změna metriky vyhodnocení, tím důležitější je funkce modelu. PFI funguje provedením této analýzy permutace napříč všemi funkcemi modelu, jeden za druhým.
V této implementaci PFI vypočítá změnu ve všech možných metrikách vyhodnocení regrese pro každou funkci a vrátí se objekty ImmutableArrayRegressionMetrics . Příklad práce s těmito výsledky pro analýzu důležitosti funkce modelu najdete v následující ukázce.
Platí pro
PermutationFeatureImportance(RankingCatalog, ITransformer, IDataView, String, String, Boolean, Nullable<Int32>, Int32)
Důležitost funkce permutace (PFI) pro řazení
public static System.Collections.Immutable.ImmutableDictionary<string,Microsoft.ML.Data.RankingMetricsStatistics> PermutationFeatureImportance (this Microsoft.ML.RankingCatalog catalog, Microsoft.ML.ITransformer model, Microsoft.ML.IDataView data, string labelColumnName = "Label", string rowGroupColumnName = "GroupId", bool useFeatureWeightFilter = false, int? numberOfExamplesToUse = default, int permutationCount = 1);
static member PermutationFeatureImportance : Microsoft.ML.RankingCatalog * Microsoft.ML.ITransformer * Microsoft.ML.IDataView * string * string * bool * Nullable<int> * int -> System.Collections.Immutable.ImmutableDictionary<string, Microsoft.ML.Data.RankingMetricsStatistics>
<Extension()>
Public Function PermutationFeatureImportance (catalog As RankingCatalog, model As ITransformer, data As IDataView, Optional labelColumnName As String = "Label", Optional rowGroupColumnName As String = "GroupId", Optional useFeatureWeightFilter As Boolean = false, Optional numberOfExamplesToUse As Nullable(Of Integer) = Nothing, Optional permutationCount As Integer = 1) As ImmutableDictionary(Of String, RankingMetricsStatistics)
Parametry
- catalog
- RankingCatalog
Katalog hodnocení.
- model
- ITransformer
Model, na kterém se má vyhodnotit důležitost funkce.
- data
- IDataView
Sada dat vyhodnocení.
- labelColumnName
- String
Název sloupce popisku Data sloupce musí být Single nebo KeyDataViewType.
- rowGroupColumnName
- String
Název sloupce GroupId
- useFeatureWeightFilter
- Boolean
Pomocí váhy funkcí můžete předfiltrovat funkce.
Omezte počet příkladů, na které se mají vyhodnotit. znamená, že se použijí až ~2 bln příklady.
- permutationCount
- Int32
Počet permutací, které se mají provést.
Návraty
Slovník mapuje každou funkci na každou funkci "příspěvky" na skóre.
Příklady
using System;
using System.Collections.Generic;
using System.Linq;
using Microsoft.ML;
namespace Samples.Dynamic.Trainers.Ranking
{
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 sample data.
var samples = GenerateData();
// 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)
.Append(mlContext.Transforms.Conversion.MapValueToKey("Label"))
.Append(mlContext.Transforms.Conversion.MapValueToKey(
"GroupId"))
.Append(mlContext.Transforms.NormalizeMinMax("Features"))
.Append(mlContext.Ranking.Trainers.FastTree());
// Fit the pipeline to the data.
var model = pipeline.Fit(data);
// Transform the dataset.
var transformedData = model.Transform(data);
// Extract the predictor.
var linearPredictor = model.LastTransformer;
// Compute the permutation metrics for the linear model using the
// normalized data.
var permutationMetrics = mlContext.Ranking.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 NDCG@1.
var sortedIndices = permutationMetrics.Select((metrics, index) => new
{
index,
metrics.NormalizedDiscountedCumulativeGains
})
.OrderByDescending(feature => Math.Abs(
feature.NormalizedDiscountedCumulativeGains[0].Mean))
.Select(feature => feature.index);
Console.WriteLine("Feature\tChange in NDCG@1\t95% Confidence in the" +
"Mean Change in NDCG@1");
var ndcg = permutationMetrics.Select(
x => x.NormalizedDiscountedCumulativeGains).ToArray();
foreach (int i in sortedIndices)
{
Console.WriteLine("{0}\t{1:G4}\t{2:G4}",
featureColumns[i],
ndcg[i][0].Mean,
1.96 * ndcg[i][0].StandardError);
}
// Expected output:
// Feature Change in NDCG@1 95% Confidence in the Mean Change in NDCG@1
// Feature2 -0.2421 0.001748
// Feature1 -0.0513 0.001184
}
private class Data
{
public float Label { get; set; }
public int GroupId { get; set; }
public float Feature1 { get; set; }
public float Feature2 { get; set; }
}
/// <summary>
/// Generate an enumerable of Data objects, creating the label as a simple
/// 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>
///
/// <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,
int groupSize = 5)
{
var rng = new Random(seed);
var max = bias + 4.5 * weight1 + 4.5 * weight2 + 0.5;
for (int i = 0; i < nExamples; i++)
{
var data = new Data
{
GroupId = i / groupSize,
Feature1 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
Feature2 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
};
// Create a noisy label.
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)
data.Label = 1;
else
data.Label = 2;
yield return data;
}
}
}
}
Poznámky
Důležitost funkce permutace (PFI) je technika k určení globální důležitosti funkcí v natrénovaného modelu strojového učení. PFI je jednoduchá, ale výkonná technika motivovaná Breimanem ve svém dokumentu Random Forest, oddíl 10 (Breiman. "Náhodné doménové struktury". Machine Learning, 2001.) Výhodou metody PFI je, že je nezávislá na modelu – funguje s jakýmkoli modelem, který lze vyhodnotit – a může používat libovolnou datovou sadu, nejen trénovací sadu, k výpočtu metrik důležitosti funkcí.
PFI funguje tak, že vezme datovou sadu označenou jako datovou sadu, zvolí funkci a probere hodnoty této funkce napříč všemi příklady, takže každý příklad teď má pro tuto funkci náhodnou hodnotu a původní hodnoty pro všechny ostatní funkce. Metrika vyhodnocení (např. NDCG) se pak vypočítá pro tuto upravenou datovou sadu a změna vyhodnocovací metriky z původní datové sady se vypočítá. Čím větší je změna metriky vyhodnocení, tím důležitější je funkce modelu. PFI funguje provedením této analýzy permutace napříč všemi funkcemi modelu, jeden za druhým.
V této implementaci PFI vypočítá změnu ve všech možných metrikách hodnocení hodnocení hodnocení pro každou funkci a ImmutableArray vrátí se objekty RankingMetrics . Příklad práce s těmito výsledky pro analýzu důležitosti funkce modelu najdete v následující ukázce.
Platí pro
PermutationFeatureImportance<TModel>(BinaryClassificationCatalog, ISingleFeaturePredictionTransformer<TModel>, IDataView, String, Boolean, Nullable<Int32>, Int32)
Důležitost funkce permutace (PFI) pro binární klasifikaci
public static System.Collections.Immutable.ImmutableArray<Microsoft.ML.Data.BinaryClassificationMetricsStatistics> PermutationFeatureImportance<TModel> (this Microsoft.ML.BinaryClassificationCatalog catalog, Microsoft.ML.ISingleFeaturePredictionTransformer<TModel> predictionTransformer, Microsoft.ML.IDataView data, string labelColumnName = "Label", bool useFeatureWeightFilter = false, int? numberOfExamplesToUse = default, int permutationCount = 1) where TModel : class;
static member PermutationFeatureImportance : Microsoft.ML.BinaryClassificationCatalog * Microsoft.ML.ISingleFeaturePredictionTransformer<'Model (requires 'Model : null)> * Microsoft.ML.IDataView * string * bool * Nullable<int> * int -> System.Collections.Immutable.ImmutableArray<Microsoft.ML.Data.BinaryClassificationMetricsStatistics> (requires 'Model : null)
<Extension()>
Public Function PermutationFeatureImportance(Of TModel As Class) (catalog As BinaryClassificationCatalog, predictionTransformer As ISingleFeaturePredictionTransformer(Of TModel), data As IDataView, Optional labelColumnName As String = "Label", Optional useFeatureWeightFilter As Boolean = false, Optional numberOfExamplesToUse As Nullable(Of Integer) = Nothing, Optional permutationCount As Integer = 1) As ImmutableArray(Of BinaryClassificationMetricsStatistics)
Parametry typu
- TModel
Parametry
- catalog
- BinaryClassificationCatalog
Katalog binární klasifikace.
- predictionTransformer
- ISingleFeaturePredictionTransformer<TModel>
Model, na kterém se má vyhodnotit důležitost funkce.
- data
- IDataView
Sada dat vyhodnocení.
- useFeatureWeightFilter
- Boolean
Pomocí váhy funkcí můžete předfiltrovat funkce.
Omezte počet příkladů, na které se mají vyhodnotit. znamená, že se použijí až ~2 bln příklady.
- permutationCount
- Int32
Počet permutací, které se mají provést.
Návraty
Pole "příspěvků" podle jednotlivých funkcí do skóre.
Příklady
using System;
using System.Collections.Generic;
using System.Linq;
using Microsoft.ML;
namespace Samples.Dynamic.Trainers.BinaryClassification
{
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 sample data.
var samples = GenerateData();
// 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)
.Append(mlContext.Transforms.NormalizeMinMax("Features"))
.Append(mlContext.BinaryClassification.Trainers
.SdcaLogisticRegression());
// Fit the pipeline to the data.
var model = pipeline.Fit(data);
// Transform the dataset.
var transformedData = model.Transform(data);
// Extract the predictor.
var linearPredictor = model.LastTransformer;
// Compute the permutation metrics for the linear model using the
// normalized data.
var permutationMetrics = mlContext.BinaryClassification
.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 AUC.
var sortedIndices = permutationMetrics
.Select((metrics, index) => new { index, metrics.AreaUnderRocCurve })
.OrderByDescending(
feature => Math.Abs(feature.AreaUnderRocCurve.Mean))
.Select(feature => feature.index);
Console.WriteLine("Feature\tModel Weight\tChange in AUC"
+ "\t95% Confidence in the Mean Change in AUC");
var auc = permutationMetrics.Select(x => x.AreaUnderRocCurve).ToArray();
foreach (int i in sortedIndices)
{
Console.WriteLine("{0}\t{1:0.00}\t{2:G4}\t{3:G4}",
featureColumns[i],
linearPredictor.Model.SubModel.Weights[i],
auc[i].Mean,
1.96 * auc[i].StandardError);
}
// Expected output:
// Feature Model Weight Change in AUC 95% Confidence in the Mean Change in AUC
// Feature2 35.15 -0.387 0.002015
// Feature1 17.94 -0.1514 0.0008963
}
private class Data
{
public bool Label { get; set; }
public float Feature1 { get; set; }
public float Feature2 { get; set; }
}
/// <summary>
/// Generate an enumerable of Data objects, creating the label as a simple
/// 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>
/// <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)
{
var rng = new Random(seed);
for (int i = 0; i < nExamples; i++)
{
var data = new Data
{
Feature1 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
Feature2 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
};
// Create a noisy label.
var value = (float)(bias + weight1 * data.Feature1 + weight2 *
data.Feature2 + rng.NextDouble() - 0.5);
data.Label = Sigmoid(value) > 0.5;
yield return data;
}
}
private static double Sigmoid(double x) => 1.0 / (1.0 + Math.Exp(-1 * x));
}
}
Poznámky
Důležitost funkce permutace (PFI) je technika k určení globální důležitosti funkcí v natrénovaného modelu strojového učení. PFI je jednoduchá, ale výkonná technika motivovaná Breimanem ve svém dokumentu Random Forest, oddíl 10 (Breiman. "Náhodné doménové struktury". Machine Learning, 2001.) Výhodou metody PFI je, že je nezávislá na modelu – funguje s jakýmkoli modelem, který lze vyhodnotit – a může používat libovolnou datovou sadu, nejen trénovací sadu, k výpočtu metrik důležitosti funkcí.
PFI funguje tak, že vezme datovou sadu označenou jako datovou sadu, zvolí funkci a probere hodnoty této funkce napříč všemi příklady, takže každý příklad teď má pro tuto funkci náhodnou hodnotu a původní hodnoty pro všechny ostatní funkce. Metrika vyhodnocení (např. AUC) se pak vypočítá pro tuto upravenou datovou sadu a změna vyhodnocovací metriky z původní datové sady se vypočítá. Čím větší je změna metriky vyhodnocení, tím důležitější je funkce modelu. PFI funguje provedením této analýzy permutace napříč všemi funkcemi modelu, jeden za druhým.
V této implementaci PFI vypočítá změnu ve všech možných metrikách vyhodnocení binární klasifikace pro každou funkci a vrátí se objekty ImmutableArrayBinaryClassificationMetrics . Příklad práce s těmito výsledky pro analýzu důležitosti funkce modelu najdete v následující ukázce.
Platí pro
PermutationFeatureImportance<TModel>(MulticlassClassificationCatalog, ISingleFeaturePredictionTransformer<TModel>, IDataView, String, Boolean, Nullable<Int32>, Int32)
Důležitost funkce permutace (PFI) pro multiclassClassification
public static System.Collections.Immutable.ImmutableArray<Microsoft.ML.Data.MulticlassClassificationMetricsStatistics> PermutationFeatureImportance<TModel> (this Microsoft.ML.MulticlassClassificationCatalog catalog, Microsoft.ML.ISingleFeaturePredictionTransformer<TModel> predictionTransformer, Microsoft.ML.IDataView data, string labelColumnName = "Label", bool useFeatureWeightFilter = false, int? numberOfExamplesToUse = default, int permutationCount = 1) where TModel : class;
static member PermutationFeatureImportance : Microsoft.ML.MulticlassClassificationCatalog * Microsoft.ML.ISingleFeaturePredictionTransformer<'Model (requires 'Model : null)> * Microsoft.ML.IDataView * string * bool * Nullable<int> * int -> System.Collections.Immutable.ImmutableArray<Microsoft.ML.Data.MulticlassClassificationMetricsStatistics> (requires 'Model : null)
<Extension()>
Public Function PermutationFeatureImportance(Of TModel As Class) (catalog As MulticlassClassificationCatalog, predictionTransformer As ISingleFeaturePredictionTransformer(Of TModel), data As IDataView, Optional labelColumnName As String = "Label", Optional useFeatureWeightFilter As Boolean = false, Optional numberOfExamplesToUse As Nullable(Of Integer) = Nothing, Optional permutationCount As Integer = 1) As ImmutableArray(Of MulticlassClassificationMetricsStatistics)
Parametry typu
- TModel
Parametry
- catalog
- MulticlassClassificationCatalog
Katalog klasifikace s více třídami.
- predictionTransformer
- ISingleFeaturePredictionTransformer<TModel>
Model, na kterém se má vyhodnotit důležitost funkce.
- data
- IDataView
Sada dat vyhodnocení.
- labelColumnName
- String
Název sloupce popisku Data sloupce musí být KeyDataViewType.
- useFeatureWeightFilter
- Boolean
Pomocí váhy funkcí můžete předfiltrovat funkce.
Omezte počet příkladů, na které se mají vyhodnotit. znamená, že se použijí až ~2 bln příklady.
- permutationCount
- Int32
Počet permutací, které se mají provést.
Návraty
Pole "příspěvků" podle jednotlivých funkcí do skóre.
Příklady
using System;
using System.Collections.Generic;
using System.Linq;
using Microsoft.ML;
namespace Samples.Dynamic.Trainers.MulticlassClassification
{
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 sample data.
var samples = GenerateData();
// 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)
.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);
// Transform the dataset.
var transformedData = model.Transform(data);
// 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 })
.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();
foreach (int i in sortedIndices)
{
Console.WriteLine("{0}\t{1:G4}\t{2:G4}",
featureColumns[i],
microAccuracy[i].Mean,
1.96 * microAccuracy[i].StandardError);
}
// Expected output:
//Feature Change in MicroAccuracy 95% Confidence in the Mean Change in MicroAccuracy
//Feature2 -0.1395 0.0006567
//Feature1 -0.05367 0.0006908
}
private class Data
{
public float Label { get; set; }
public float Feature1 { get; set; }
public float Feature2 { get; set; }
}
/// <summary>
/// Generate an enumerable of Data objects, creating the label as a simple
/// 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>
/// <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)
{
var rng = new Random(seed);
var max = bias + 4.5 * weight1 + 4.5 * weight2 + 0.5;
for (int i = 0; i < nExamples; i++)
{
var data = new Data
{
Feature1 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
Feature2 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
};
// Create a noisy label.
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)
data.Label = 1;
else
data.Label = 2;
yield return data;
}
}
}
}
Poznámky
Důležitost funkce permutace (PFI) je technika k určení globální důležitosti funkcí v natrénovaného modelu strojového učení. PFI je jednoduchá, ale výkonná technika motivovaná Breimanem ve svém dokumentu Random Forest, oddíl 10 (Breiman. "Náhodné doménové struktury". Machine Learning, 2001.) Výhodou metody PFI je, že je nezávislá na modelu – funguje s jakýmkoli modelem, který lze vyhodnotit – a může používat libovolnou datovou sadu, nejen trénovací sadu, k výpočtu metrik důležitosti funkcí.
PFI funguje tak, že vezme datovou sadu označenou jako datovou sadu, zvolí funkci a probere hodnoty této funkce napříč všemi příklady, takže každý příklad teď má pro tuto funkci náhodnou hodnotu a původní hodnoty pro všechny ostatní funkce. Metrika vyhodnocení (např. mikropřesnost) se pak vypočítá pro tuto upravenou datovou sadu a změna vyhodnocovací metriky z původní datové sady se vypočítá. Čím větší je změna metriky vyhodnocení, tím důležitější je funkce modelu. PFI funguje provedením této analýzy permutace napříč všemi funkcemi modelu, jeden za druhým.
V této implementaci PFI vypočítá změnu ve všech možných metrikách vyhodnocení klasifikace více tříd pro každou funkci a ImmutableArray vrátí se objekty MulticlassClassificationMetrics . Příklad práce s těmito výsledky pro analýzu důležitosti funkce modelu najdete v následující ukázce.
Platí pro
PermutationFeatureImportance<TModel>(RegressionCatalog, ISingleFeaturePredictionTransformer<TModel>, IDataView, String, Boolean, Nullable<Int32>, Int32)
Důležitost funkce permutace (PFI) pro regresi
public static System.Collections.Immutable.ImmutableArray<Microsoft.ML.Data.RegressionMetricsStatistics> PermutationFeatureImportance<TModel> (this Microsoft.ML.RegressionCatalog catalog, Microsoft.ML.ISingleFeaturePredictionTransformer<TModel> predictionTransformer, Microsoft.ML.IDataView data, string labelColumnName = "Label", bool useFeatureWeightFilter = false, int? numberOfExamplesToUse = default, int permutationCount = 1) where TModel : class;
static member PermutationFeatureImportance : Microsoft.ML.RegressionCatalog * Microsoft.ML.ISingleFeaturePredictionTransformer<'Model (requires 'Model : null)> * Microsoft.ML.IDataView * string * bool * Nullable<int> * int -> System.Collections.Immutable.ImmutableArray<Microsoft.ML.Data.RegressionMetricsStatistics> (requires 'Model : null)
<Extension()>
Public Function PermutationFeatureImportance(Of TModel As Class) (catalog As RegressionCatalog, predictionTransformer As ISingleFeaturePredictionTransformer(Of TModel), data As IDataView, Optional labelColumnName As String = "Label", Optional useFeatureWeightFilter As Boolean = false, Optional numberOfExamplesToUse As Nullable(Of Integer) = Nothing, Optional permutationCount As Integer = 1) As ImmutableArray(Of RegressionMetricsStatistics)
Parametry typu
- TModel
Parametry
- catalog
- RegressionCatalog
Regresní katalog.
- predictionTransformer
- ISingleFeaturePredictionTransformer<TModel>
Model, na kterém se má vyhodnotit důležitost funkce.
- data
- IDataView
Sada dat vyhodnocení.
- useFeatureWeightFilter
- Boolean
Pomocí váhy funkcí můžete předfiltrovat funkce.
Omezte počet příkladů, na které se mají vyhodnotit. znamená, že se použijí až ~2 bln příklady.
- permutationCount
- Int32
Počet permutací, které se mají provést.
Návraty
Pole "příspěvků" podle jednotlivých funkcí do skóre.
Příklady
using System;
using System.Collections.Generic;
using System.Linq;
using Microsoft.ML;
namespace Samples.Dynamic.Trainers.Regression
{
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 sample data.
var samples = GenerateData();
// 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)
.Append(mlContext.Transforms.NormalizeMinMax("Features"))
.Append(mlContext.Regression.Trainers.Ols());
// Fit the pipeline to the data.
var model = pipeline.Fit(data);
// Transform the dataset.
var transformedData = model.Transform(data);
// Extract the predictor.
var linearPredictor = model.LastTransformer;
// Compute the permutation metrics for the linear model using the
// normalized data.
var permutationMetrics = mlContext.Regression
.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 RMSE.
var sortedIndices = permutationMetrics
.Select((metrics, index) => new
{
index,
metrics.RootMeanSquaredError
})
.OrderByDescending(feature => Math.Abs(
feature.RootMeanSquaredError.Mean))
.Select(feature => feature.index);
Console.WriteLine("Feature\tModel Weight\tChange in RMSE\t95%" +
"Confidence in the Mean Change in RMSE");
var rmse = permutationMetrics.Select(x => x.RootMeanSquaredError)
.ToArray();
foreach (int i in sortedIndices)
{
Console.WriteLine("{0}\t{1:0.00}\t{2:G4}\t{3:G4}",
featureColumns[i],
linearPredictor.Model.Weights[i],
rmse[i].Mean,
1.96 * rmse[i].StandardError);
}
// Expected output:
// Feature Model Weight Change in RMSE 95% Confidence in the Mean Change in RMSE
// Feature2 9.00 4.009 0.008304
// Feature1 4.48 1.901 0.003351
}
private class Data
{
public float Label { get; set; }
public float Feature1 { get; set; }
public float Feature2 { get; set; }
}
/// <summary>
/// Generate an enumerable of Data objects, creating the label as a simple
/// 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>
/// <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)
{
var rng = new Random(seed);
for (int i = 0; i < nExamples; i++)
{
var data = new Data
{
Feature1 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
Feature2 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
};
// Create a noisy label.
data.Label = (float)(bias + weight1 * data.Feature1 + weight2 *
data.Feature2 + rng.NextDouble() - 0.5);
yield return data;
}
}
}
}
Poznámky
Důležitost funkce permutace (PFI) je technika k určení globální důležitosti funkcí v natrénovaného modelu strojového učení. PFI je jednoduchá, ale výkonná technika motivovaná Breimanem ve svém dokumentu Random Forest, oddíl 10 (Breiman. "Náhodné doménové struktury". Machine Learning, 2001.) Výhodou metody PFI je, že je nezávislá na modelu – funguje s jakýmkoli modelem, který lze vyhodnotit – a může používat libovolnou datovou sadu, nejen trénovací sadu, k výpočtu metrik důležitosti funkcí.
PFI funguje tak, že vezme datovou sadu označenou jako datovou sadu, zvolí funkci a probere hodnoty této funkce napříč všemi příklady, takže každý příklad teď má pro tuto funkci náhodnou hodnotu a původní hodnoty pro všechny ostatní funkce. Metrika vyhodnocení (např. R-squared) se pak vypočítá pro tuto upravenou datovou sadu a změna vyhodnocovací metriky z původní datové sady se vypočítá. Čím větší je změna metriky vyhodnocení, tím důležitější je funkce modelu. PFI funguje provedením této analýzy permutace napříč všemi funkcemi modelu, jeden za druhým.
V této implementaci PFI vypočítá změnu ve všech možných metrikách vyhodnocení regrese pro každou funkci a vrátí se objekty ImmutableArrayRegressionMetrics . Příklad práce s těmito výsledky pro analýzu důležitosti funkce modelu najdete v následující ukázce.
Platí pro
PermutationFeatureImportance<TModel>(RankingCatalog, ISingleFeaturePredictionTransformer<TModel>, IDataView, String, String, Boolean, Nullable<Int32>, Int32)
Důležitost funkce permutace (PFI) pro řazení
public static System.Collections.Immutable.ImmutableArray<Microsoft.ML.Data.RankingMetricsStatistics> PermutationFeatureImportance<TModel> (this Microsoft.ML.RankingCatalog catalog, Microsoft.ML.ISingleFeaturePredictionTransformer<TModel> predictionTransformer, Microsoft.ML.IDataView data, string labelColumnName = "Label", string rowGroupColumnName = "GroupId", bool useFeatureWeightFilter = false, int? numberOfExamplesToUse = default, int permutationCount = 1) where TModel : class;
static member PermutationFeatureImportance : Microsoft.ML.RankingCatalog * Microsoft.ML.ISingleFeaturePredictionTransformer<'Model (requires 'Model : null)> * Microsoft.ML.IDataView * string * string * bool * Nullable<int> * int -> System.Collections.Immutable.ImmutableArray<Microsoft.ML.Data.RankingMetricsStatistics> (requires 'Model : null)
<Extension()>
Public Function PermutationFeatureImportance(Of TModel As Class) (catalog As RankingCatalog, predictionTransformer As ISingleFeaturePredictionTransformer(Of TModel), data As IDataView, Optional labelColumnName As String = "Label", Optional rowGroupColumnName As String = "GroupId", Optional useFeatureWeightFilter As Boolean = false, Optional numberOfExamplesToUse As Nullable(Of Integer) = Nothing, Optional permutationCount As Integer = 1) As ImmutableArray(Of RankingMetricsStatistics)
Parametry typu
- TModel
Parametry
- catalog
- RankingCatalog
Katalog hodnocení.
- predictionTransformer
- ISingleFeaturePredictionTransformer<TModel>
Model, na kterém se má vyhodnotit důležitost funkce.
- data
- IDataView
Sada dat vyhodnocení.
- labelColumnName
- String
Název sloupce popisku Data sloupce musí být Single nebo KeyDataViewType.
- rowGroupColumnName
- String
Název sloupce GroupId
- useFeatureWeightFilter
- Boolean
Pomocí váhy funkcí můžete předfiltrovat funkce.
Omezte počet příkladů, na které se mají vyhodnotit. znamená, že se použijí až ~2 bln příklady.
- permutationCount
- Int32
Počet permutací, které se mají provést.
Návraty
Pole "příspěvků" podle jednotlivých funkcí do skóre.
Příklady
using System;
using System.Collections.Generic;
using System.Linq;
using Microsoft.ML;
namespace Samples.Dynamic.Trainers.Ranking
{
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 sample data.
var samples = GenerateData();
// 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)
.Append(mlContext.Transforms.Conversion.MapValueToKey("Label"))
.Append(mlContext.Transforms.Conversion.MapValueToKey(
"GroupId"))
.Append(mlContext.Transforms.NormalizeMinMax("Features"))
.Append(mlContext.Ranking.Trainers.FastTree());
// Fit the pipeline to the data.
var model = pipeline.Fit(data);
// Transform the dataset.
var transformedData = model.Transform(data);
// Extract the predictor.
var linearPredictor = model.LastTransformer;
// Compute the permutation metrics for the linear model using the
// normalized data.
var permutationMetrics = mlContext.Ranking.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 NDCG@1.
var sortedIndices = permutationMetrics.Select((metrics, index) => new
{
index,
metrics.NormalizedDiscountedCumulativeGains
})
.OrderByDescending(feature => Math.Abs(
feature.NormalizedDiscountedCumulativeGains[0].Mean))
.Select(feature => feature.index);
Console.WriteLine("Feature\tChange in NDCG@1\t95% Confidence in the" +
"Mean Change in NDCG@1");
var ndcg = permutationMetrics.Select(
x => x.NormalizedDiscountedCumulativeGains).ToArray();
foreach (int i in sortedIndices)
{
Console.WriteLine("{0}\t{1:G4}\t{2:G4}",
featureColumns[i],
ndcg[i][0].Mean,
1.96 * ndcg[i][0].StandardError);
}
// Expected output:
// Feature Change in NDCG@1 95% Confidence in the Mean Change in NDCG@1
// Feature2 -0.2421 0.001748
// Feature1 -0.0513 0.001184
}
private class Data
{
public float Label { get; set; }
public int GroupId { get; set; }
public float Feature1 { get; set; }
public float Feature2 { get; set; }
}
/// <summary>
/// Generate an enumerable of Data objects, creating the label as a simple
/// 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>
///
/// <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,
int groupSize = 5)
{
var rng = new Random(seed);
var max = bias + 4.5 * weight1 + 4.5 * weight2 + 0.5;
for (int i = 0; i < nExamples; i++)
{
var data = new Data
{
GroupId = i / groupSize,
Feature1 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
Feature2 = (float)(rng.Next(10) * (rng.NextDouble() - 0.5)),
};
// Create a noisy label.
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)
data.Label = 1;
else
data.Label = 2;
yield return data;
}
}
}
}
Poznámky
Důležitost funkce permutace (PFI) je technika k určení globální důležitosti funkcí v natrénovaného modelu strojového učení. PFI je jednoduchá, ale výkonná technika motivovaná Breimanem ve svém dokumentu Random Forest, oddíl 10 (Breiman. "Náhodné doménové struktury". Machine Learning, 2001.) Výhodou metody PFI je, že je nezávislá na modelu – funguje s jakýmkoli modelem, který lze vyhodnotit – a může používat libovolnou datovou sadu, nejen trénovací sadu, k výpočtu metrik důležitosti funkcí.
PFI funguje tak, že vezme datovou sadu označenou jako datovou sadu, zvolí funkci a probere hodnoty této funkce napříč všemi příklady, takže každý příklad teď má pro tuto funkci náhodnou hodnotu a původní hodnoty pro všechny ostatní funkce. Metrika vyhodnocení (např. NDCG) se pak vypočítá pro tuto upravenou datovou sadu a změna vyhodnocovací metriky z původní datové sady se vypočítá. Čím větší je změna metriky vyhodnocení, tím důležitější je funkce modelu. PFI funguje provedením této analýzy permutace napříč všemi funkcemi modelu, jeden za druhým.
V této implementaci PFI vypočítá změnu ve všech možných metrikách hodnocení hodnocení hodnocení pro každou funkci a ImmutableArray vrátí se objekty RankingMetrics . Příklad práce s těmito výsledky pro analýzu důležitosti funkce modelu najdete v následující ukázce.