Dela via

RankingExperimentResultExtensions.Best Method

  • Reference

Definition

Namespace:
Microsoft.ML.AutoML
Assembly:
Microsoft.ML.AutoML.dll
Package:
Microsoft.ML.AutoML v0.21.1

Overloads

Best(IEnumerable<CrossValidationRunDetail<RankingMetrics>>, RankingMetric, UInt32)

Select the best run from an enumeration of experiment cross validation runs.
Best(IEnumerable<RunDetail<RankingMetrics>>, RankingMetric, UInt32)

Select the best run from an enumeration of experiment runs.

Best(IEnumerable<CrossValidationRunDetail<RankingMetrics>>, RankingMetric, UInt32)

Select the best run from an enumeration of experiment cross validation runs.

public static Microsoft.ML.AutoML.CrossValidationRunDetail<Microsoft.ML.Data.RankingMetrics> Best (this System.Collections.Generic.IEnumerable<Microsoft.ML.AutoML.CrossValidationRunDetail<Microsoft.ML.Data.RankingMetrics>> results, Microsoft.ML.AutoML.RankingMetric metric = Microsoft.ML.AutoML.RankingMetric.Ndcg, uint optimizationMetricTruncationLevel = 10);
static member Best : seq<Microsoft.ML.AutoML.CrossValidationRunDetail<Microsoft.ML.Data.RankingMetrics>> * Microsoft.ML.AutoML.RankingMetric * uint32 -> Microsoft.ML.AutoML.CrossValidationRunDetail<Microsoft.ML.Data.RankingMetrics>
<Extension()>
Public Function Best (results As IEnumerable(Of CrossValidationRunDetail(Of RankingMetrics)), Optional metric As RankingMetric = Microsoft.ML.AutoML.RankingMetric.Ndcg, Optional optimizationMetricTruncationLevel As UInteger = 10) As CrossValidationRunDetail(Of RankingMetrics)

Parameters

results
IEnumerable<CrossValidationRunDetail<RankingMetrics>>

Enumeration of AutoML experiment cross validation run results.

metric
RankingMetric

Metric to consider when selecting the best run.

optimizationMetricTruncationLevel
UInt32

Maximum truncation level for computing (N)DCG. Defaults to 10.

Returns

CrossValidationRunDetail<RankingMetrics>

The best experiment run.

Applies to

Best(IEnumerable<RunDetail<RankingMetrics>>, RankingMetric, UInt32)

Select the best run from an enumeration of experiment runs.

public static Microsoft.ML.AutoML.RunDetail<Microsoft.ML.Data.RankingMetrics> Best (this System.Collections.Generic.IEnumerable<Microsoft.ML.AutoML.RunDetail<Microsoft.ML.Data.RankingMetrics>> results, Microsoft.ML.AutoML.RankingMetric metric = Microsoft.ML.AutoML.RankingMetric.Ndcg, uint optimizationMetricTruncationLevel = 10);
static member Best : seq<Microsoft.ML.AutoML.RunDetail<Microsoft.ML.Data.RankingMetrics>> * Microsoft.ML.AutoML.RankingMetric * uint32 -> Microsoft.ML.AutoML.RunDetail<Microsoft.ML.Data.RankingMetrics>
<Extension()>
Public Function Best (results As IEnumerable(Of RunDetail(Of RankingMetrics)), Optional metric As RankingMetric = Microsoft.ML.AutoML.RankingMetric.Ndcg, Optional optimizationMetricTruncationLevel As UInteger = 10) As RunDetail(Of RankingMetrics)

Parameters

results
IEnumerable<RunDetail<RankingMetrics>>

Enumeration of AutoML experiment run results.

metric
RankingMetric

Metric to consider when selecting the best run.

optimizationMetricTruncationLevel
UInt32

Maximum truncation level for computing (N)DCG. Defaults to 10.

Returns

RunDetail<RankingMetrics>

The best experiment run.

Applies to